2018 International Conference on Innovation and Intelligence for Informatics, Computing, and Technologies (3ICT) Program
Sunday, November 18
Sunday, November 18 8:00 - 9:00
R1: Registration-Day1
Room: S45-Zain E-Learning Center
Sunday, November 18 9:00 - 9:10
OC-1: Opening Ceremony
Room: S45-101-Zain E-Learning Center
Chair: Faisal Hammad
Sunday, November 18 9:10 - 9:20
OC-2: Talk by His Excellency President of the University of Bahrain
Prof. Riyad Hamzah
Room: S45-101-Zain E-Learning CenterSunday, November 18 9:20 - 9:40
KS-1: Keynote Speaker
Be Careful: I can Read your Mind
Prof. Magdy A. Bayoumi, University of Louisiana at Lafayette, USA
Room: S45-101-Zain E-Learning CenterThis is one of the ultimate goals of the "Brain Initiative" which may be real by 2030. In General, to be able to control our surrounding world by the power of our mind, to control the artificial limbs by our thoughts, and to read people's brain is a dream and it is an exciting one.Brain Computer/Machine Interface (BCI/BMI) is one of the enabling technology towards these goals. We have designed and developed BCI chips that can overcome the main challenges of interfacing with human brain, mainly: low bandwidth communication, small chip area, low power and low heat dissipation, and tolerant to noise. This new chip is based on spike sorting. Probably, the more reflective title of this talk is "Brain on Silicon" Finally, a case study of early prediction, warning, and detection of seizure using the developed chip will be illustrated.
Sunday, November 18 9:40 - 9:50
OC-3: Appreciation for Keynote Speakers
Room: S45-101-Zain E-Learning Center
Sunday, November 18 9:50 - 10:00
SB-1: Short Break
Room: S45-Zain E-Learning Center
Sunday, November 18 10:00 - 11:40
S1: Smart Cities
Room: S45-108-Zain E-Learning Cente
Chairs: Princy Johnson, Hayat Mohamed Abdulla Ali
- 10:00 Dynamic Model of Soil Moisture for Smart Irrigation Systems
-
In geographical areas subject to an arid climate, one of the most precious resources is water, which often get wasted due to inefficient irrigation systems. Automated irrigation systems which relies on closed loop control and feedback from sensors can help to closely match the water supply to the crop demand and avoid waste. In recent years, predictive control strategies were proposed for high tech watering systems which rely heavily on complex models incorporating variables such as crop water demand, soil evapotranspiration, weather conditions etc.. Through these models it is possible to anticipate the water demand of the crop, and give optimal irrigation scheduling with a significant reduction in water consumption. In this paper we developed a model of the soil moisture dynamics based on climatological data of the Kingdom of Bahrain. The model, which is based on the hydraulic balance approach, was validated by comparing its estimated output with real measurements
- 10:20 BIM from Conceptual Model to Construction
-
The paper investigates the architectural models and the construction process while using Building Information Modelling. Building BIM model becomes vital for the designers in conceptual designing; collaborative design and communication between all designers are essential during these design phases. Multidisciplinary model is the first step for the designers of different disciplines to communicate and collaborate in such matter. After creating the multidisciplinary model in conceptual designing, is it enough to use into the construction process? Are there more details that should be added into the model? The paper attempts at shed more light on the obstacles faced and functions needed to use the conceptual model of BIM into the construction process. The paper explores the details required to the BIM models to be utilised into the construction process. A main methodological tool is applied by synthesizing and extracting from the literature review of related research areas. The secondary data lead to certain categories of BIM details that have to be applied by the BIM users after the conceptual design phases. The paper concludes to solid results and certain recommendations during creating BIM model.
- 10:40 Smart City: An Investigation Toward a Proposed Maturity Model
-
Smart City is an emerging strategy to solve the problems resulted from the increase in urban population and rapid urbanization. The global market for smart city solutions, services, and investments in the world is expected to be very high in 2026. However, there are very few scientific papers that propose a Smart City maturity model which provide directions or support such investment decisions. Due to the complexity and limited experience concerning Smart City within countries, there is very few researches that can be used to evaluate the city's achievements in implementing the smart city initiatives. Also, there is a limitation in measuring the level of maturity (as is) and determining the next future level (to be). The main objective of this paper is to identify and assess the current maturity models and develop a Customized Smart City Maturity Model for Kingdom of Bahrain. Furthermore, it aims to highlight the status of Smart City in the Kingdom and assess the level of maturity in one of Smart Cities' dimension (Smart Governance) using the customize maturity model.
- 11:00 A Low-cost Solar Farm Monitoring System Based on Cloud Database
-
Renewable energies have become important sources of electricity generation in modern times, with solar energy being one of the best types of such energy since solar radiation can be converted directly into electricity. To harness the solar energy in the best way, specialists in this field resort to solar farms where a large number of solar panels are installed in an open area and the electricity produced is being transferred into towns. The design of a reliable monitoring system for such a large number of solar panels becomes the essential part of such design to reach the needed efficiency. This paper introduces a low-cost practical implementation of a monitoring system for an existing solar system installed on the building of the Ministry of Science and Technology (MST) in Baghdad based on Wireless Sensor Network (WSN) technique. The "ThingSpeak" cloud database is utilized to store and retrieve the monitored data and the PHP and HTML programming languages are used to design a Graphical User Interface (GUI) to display these data.
- 11:20 Smart Transportation System: An Investigation of the security challenges
-
The world today is moving toward the connected environment because of the advance in the information technology, smart cities is a result of this evolution. One of the important technologies that get high attention within the framework of the smart cities is the smart transportation. The smart transportation or the intelligent transportation system (ITS) solved many issues related to the transportation. However, it might be exposed to many security and privacy issues. Therefore, this research aims to investigate the security issues, privacy issues and the solutions for the smart transportation. Furthermore, the contribution of this research will promote the researchers to the different types of the security and privacy issues with the solutions so that they can use the outcomes as a base for their further researches.
S2: Telecommunication and Networks-1
Room: S45-06-Zain E-Learning Center
Chairs: Abdulnasir Hossen, Reham Almesaeed
- 10:00 A Comparison Analysis of Fault Detection Algorithms in Wireless Sensor Networks
-
Sensors are capable of performing different tasks such as receiving various environmental information and have generated the idea of creating and expanding the traditional networks to Wireless Sensor Networks (WSN). A WSN include various sensor nodes which are distributed in the environment to gather data. The sensor node's location is not necessarily defined and not specified. Such a feature allows us to release them in hazardous or inaccessible places. On the other hand, this means that the WSN protocols and algorithms must have the ability to self-deny. Other unique features of WSNs are the ability of sensor nodes to collaborate and coordinate with each other. A processor is assigned to each sensor node which initially performs a series of acquired simple information processing to transmits the semi-processed data instead of sending all raw data to the center, Base Station (BS) or the sink node. Fault detections in WSN are of particular importance due to unique technical and conceptual challenges. In this paper, due to the specific limitations and operating conditions of wireless sensor networks, different fault detection algorithms were examined. The best way to improve the ability of fault detection is to reduce energy consumption in sensor nodes. In such techniques, using the residue number system in the wireless sensor network structure improves repair and fault detection. Moreover, modules are attached whenever we require distinguishing and correcting the information security in the number system. Although there are various algorithms yet the Residue Number System algorithm has higher performance.
- 10:20 Software Defined-Network for Real-time Operations of Business Intelligence Tasks
-
Information and Communication Technology (ICT) is a crucial tool for Business Intelligence. ICT tools help in an operation that yields a "real-time reporting with analytical alert" through the use of BI operation. The efficient handle of the flow of data from one node to the other by BI operation required many network resources, limited resource is a drawback to BI tasks that need a real-time operation. If bandwidth is shared among applications that use the internet. It is consider as one of the crucial obstacles to BI operations. Therefore, a dynamic tasks scheduling application is required. It should be able to manage the flow of data in a network involving of BI operation. For that reason, this paper proposes an effective routing scheme required for BI tasks in order to provide real-time reporting with the analytical decision without link congestion for any task. The proposed mechanism can be used to control the actions and the forwarding rules in the network. This increases the efficiency of a network by rerouting traffics between links. It utilizes the idle links and directs the traffic from and to them to raise the network total efficiency rate. Consequently, bandwidths will be remarkably higher and the link congestion problem will be avoided from that network. In the experiments of emulating congested traffic, the paper gets better network performance of throughput, jitter, and packet loss rate using the proposed traffic rerouting, as opposed to the environment without traffic rerouting.
- 10:40 Crosstalk Noise Controller in WDM Optical Network on Chip Routers
-
The computer applications performance growing exponentially and they are greedier in computing resources. To satisfy these demands the multiprocessor system-on-chip (MPSoC) is an attractive solution which became one of the important technologies in our days. However, the communication between the different processors' cores presents the first challenge front the high performance of MPSoC. Network on chip (NoC) is one of the promoting solution to improve on chip communication between cores. Nevertheless, NoC potential limited by physical limitation, power consumption, latency and bandwidth in the both case: increasing data exchange or scalability of multicores. To avoid these limitations, the optical communication can be a perfect solution which it offers a wider bandwidth and lower power consumption. The silicon photonics technology proposes many optical devices on chip, based on a new technology named Optical Network-on-Chip (ONoC) has been introduced in MPSoC. However, the factoring nature of the ONoC components induces a malicious impairment called crosstalk noise. This serious problem deteriorates the quality of signals and degrades network performance. Many types of optical switches are developed to ONoC which they present the backbone of the network and the first resource of the crosstalk noise. As a result, detect and monitor the crosstalk noise becoming a challenge to keep the performance in the ONoC. In this article, we propose a new hardware solution to detect and control the crosstalk noise in ONoC components. Particularly, we present the different optical devices presented in ONoC and an analytic model of the crosstalk noise at the optical devices. The proposed hardware system is presented and implemented in objective to test his performance and efficiency in terms of execution time, area occupation and scalability. Moreover our system reach the high performance with 23 ms in execution time
- 11:00 Tracking High-Speed Users Using SNR-CQI Mapping in LTE-A Networks
-
The task of estimating, tracking and controlling the speed of user equipment (UE), particularly with speed above allowed limits, is a challenge which has a significant impact on radio resource and mobility management in Long Term Evolution (LTE). In this paper, we propose an algorithm that relies on transmitting Channel Quality Indicator CQI bits through LTE network's remote IP Multimedia Subsystem (IMS) application server to analyze and to map the UE speed and to determine if the UE is above the speed limit. The paper presents a new approach to estimate, track, and control users moving above speed limits in LTE-Advanced (LTE-A) networks using mapping of the uplink CQI index of the UE since the CQI range can provide an indication to the system regarding the movement and the speed of the UE.
- 11:20 Solving the Discrete Cost Multicommodity Network Design Problem to Optimality
-
Multicommodity Network Design problems arise in the strategic and tactical planning processes and have many applications mainly in the fields of telecommunication and logistics. Thus, solving this challenging NP-hard problem is crucial for the profitable business of network operators. In this work, we focus on the Discrete Cost Multicommodity Network Design Problem (DCMNDP), with multiple discrete facilities (corresponding to different technologies) to be installed on the edges. Each facility is bidirectional and has known discrete capacity and cost. For this problem, we present a tailored Benders decomposition using an arc-flow based formulation and we strengthen it by adding some valid inequalities. Preliminary computational results conducted on both real-world instances and randomly generated instances are reported.
S3: Artificial Intelligence, Robotics, and Data Mining
Room: S45-11-Zain E-Learning Center
Chairs: Noora Alghatam, Alrence S Halibas
- 10:00 Application Of Artificial Neural Networks For Exploratory Analysis Of Small Dataset
-
This study aims to explore the use of Artificial Neural Networks (ANNs) for estimating the relationship between accidents and other variables with a small dataset. Analysis of road traffic accidents is often hampered due to insufficient datasets. Especially, for the cases when specific highway facilities are considered. This issue is also gaining importance for analysing traveller behaviour with the advent of new technologies and implementation of concepts of smart cities. The accident sites selected for this study comprise of unsignalized intersections in Penang State of Malaysia. Accidents in Malaysia have become a major concern for the authorities. However, the data collection is a major issue hindering its analysis because of limited datasets. The safe operation of traffic on unsignalized intersections mainly depends on drivers' judgement and decision making. Hence, the safety considerations on such locations are peculiar in comparison to other facilities. These facts led to carrying out the study for these sites. Two types of ANNs were used i.e. Multilayer Feedforward (MLFF) and linear. In addition, regression model and Mann-Whitney test were also used to compare the results from ANNs. It was found that regression model as well as Mann-Whitney test gave inconclusive results for the available dataset. On the other hand, ANNs were able to approximate the relationship in conformity to the previous studies. However, larger datasets are expected to give better accuracies for regression as well as ANNs.
- 10:20 Execution Time Performance For Different Variants of R-Eclat Algorithm
-
Association rules aim to mining interesting correlations and association among itemsets in the transaction databases. The relationship between itemsets can be found as frequent pattern and infrequent pattern. The mining infrequent pattern has been an interesting issue in discovering the uninteresting patterns which are rarely occurring. However, rarely itemsets occurrences may provide valuable information in knowledge mining process. R-Eclat is a novel algorithm that uses in determining infrequent itemsets in a transaction database. This paper proposed IF-Postdiffset as a new format variant for R-Eclat algorithm. Besides, the issue of discovering infrequent itemsets mining from the transactional database via different format variants of the R-Eclat algorithm will be discussed. Experimental results will show the efficiency and effectiveness of the proposed approach.
- 10:40 ONTO SQAM - A Model for Analyzing Seafood Quality Based on Ontology
-
One of the most precious exporting sectors in the Indian economy is the seafood industry. Sustained measures to ensure the quality of export materials are very critical to further increase the growth in seafood export. To ensure the overall quality of seafood, a knowledge-based system entitled 'Ontology Based Quality Analyzer and Miner (ONTO SQAM) is developed. Ontologies are used to represent the seafood domain. Different algorithms are suggested for analysis, mining, and prediction to ensure the seafood quality. The key features of the model are the development of algorithms to generate frequent patterns of quality seafood and to predict the success rates of seafood based on the catching center. Indian Seafood industry is taken as the case study in this research. This paper presents only the architecture of the system. The implementation of the model will be discussed in an extended version of this paper.
- 11:00 Multi-Human-Multi-Robot Facial Interaction System
-
Recent advancement in the service and manufacturing industry has increased the productivity of the process as well as the quality of the products. In order to improve it to a more sophisticated level, robot is the necessary element of the service and manufacturing industry. To give a more advance shape, the robots must have to interact with one another and with the humans. In this research, a system has been designed and developed in which multiple human and multiple robot will be able to interact with one another using facial features directions. The system includes the face detection, facial features detections and classification of features directions, so that the human can give specific facial features direction to a specific robot for a specific type of action. For face detection Viola Jones method has been used, for facial features detection modified version of Viola Jones methods is used. After the facial features detection for facial features direction classification in order to command the robot convolutional neural network (CNN) architecture Resnet50 is used.
- 11:20 Features-Level Software Effort Estimation Using Machine Learning Algorithms
-
Software effort estimation is a paramount mission in the software development process, which covered by project managers and software engineers. In the early stages, software system features are the only available measures. Therefore, cost estimation is a mission that comes under the planning stage of software venture management. In this paper, various machine learning algorithms are used to build software effort estimation models from software features. Artificial Neural Network (ANN), Support Vector Machines (SVM), K-star, and Linear Regression machine learning algorithms are evaluated on a public dataset with actual software efforts. Results showed that machine learning approach can be dependable on predicting the future effort of a software system.
Sunday, November 18 11:40 - 12:00
P1: Prayer
Room: S45-Zain E-Learning Center
Sunday, November 18 12:00 - 12:20
KS-2: Keynote Speaker
Enabling Collaboration in Internet of Things Environments: Security, Trust and Resilience
Prof. Mohamed Eltoweissy, Virginia Military Institute, USA
Room: S45-101-Zain E-Learning CenterChair: Noora Alghatam
Recently, Cyberspace has been designated as the fifth, and only human-made, dimension of operations. Cyberspace is becoming the nervous system of our modern society as it is increasingly being integrated in virtually all aspects of control and communication. Today's Cyberspace comprises highly dynamic, interdependent global network of information technology infrastructures, telecommunications networks, computing systems, integrated sensors, control systems, embedded processors and controllers giving rise to the pervasive "Internet of Things (IOT)." The complexity and heterogeneity of the IoT systems coupled with evolving advanced persistent threats as the new reality, call for platforms to manage security, trust and resilience as services. This talk will focus on how to enable effective collaboration in IOT-powered environments. Specifically, it will discuss IOT security, trust and resilience as essential enablers for effective collaboration. The talk will outline some challenges and proposes a platform that would provide a comprehensive approach for enabling collaboration.
Sunday, November 18 12:20 - 12:40
KS-3: Keynote Speaker
IoT and Consumer Healthcare- Opportunities and Challenges
Dr. Chitra Balakrishna, program Leader in cyber security at Edge Hill University, United Kingdom
Room: S45-101-Zain E-Learning CenterChair: Aisha Bushager
The term healthcare has a very wide scope that ranges from lifestyle and wellness right up to care for acute conditions. In the recent past, a large number of devices enabled by IoT have become available in the market, that can monitor various aspects of lifestyle and biological functions. As a consequence, the potential for obtaining detailed data on life style, habits and behaviour of an individual exists. Such data provides a feedback to an individual for compliance with ``healthy guidelines'' as well as contribute information to the healthcare provider for use in diagnosis, in the event of an ailment.The talk addresses various aspects of care that can benefit from IoT-enabled consumer grade health monitoring devices, various opportunities and challenges in this ‘Consumer Healthcare' landscape.
Sunday, November 18 12:40 - 1:00
SB-2: Short Break
Room: S45-Zain E-Learning Center
Sunday, November 18 1:00 - 4:50
PS-1: Poster Session
Room: S45-Zain E-Learning Center
Sunday, November 18 1:00 - 2:20
S6: From RFID to Cyber-Physical Systems: Reality, Dreams, and Fantasy
IEEE Invited Speaker - This talk is sponsored by IEEE council on RFID
Distinguished Lecturer: Prof. Magdy A. Bayoumi, University of Louisiana at Lafayette, USA
Room: S45-108-Zain E-Learning CenteChairs: Isa Salman Qamber, Adel Abdulla
S5: Informatics-1 (e-Government Services)
Room: S45-06-Zain E-Learning Center
Chairs: Abdul Fattah Salman, Wael Abdelhameed
- 1:00 The Cultural Impact on User Interface Design: The Case of e-Government of Kingdom of Bahrain
-
The internet and web services continue to grow rapidly and so does the needs of users. Users prefer easy and fast services that can be accessed everywhere such as paying utility bills and scheduling appointments.In doing so, on-line services need to be provided by governments to facilitate easy access.In this age of globalization, we live in a highly diversified world in which people from different cultural backgrounds live together.Users interactions become more appealing if cultural characteristics are incorporated into the interface design. Culture could have an important impact upon people using the web services. Thus, designing web services for a multi-cultural society will require studying in depth the effect of the society cultural characteristics. The user interface can be designed using those characteristics that help in gaining the user's attention. Previous studies indicate that users are more comfortable when interacting with a user interface related to their own culture. This paper investigates the influence of culture on the user interface design of Bahrain e-Government services to examine the views of cultural effect on user interface design. The results show the positive and negative impact of users culture on the interface design.
- 1:20 An Evaluation of Technological and Physiological Barriers in E-Commerce Adoption in the British SME's
-
Electronic commerce in the recent decades has been adopted by more and more companies or sectors to aid in their growth. It is obvious that the concept of e-commerce has been frequently discussed over the recent decades of fast development of internet technologies. Led by some giant companies of internet, the e-commerce industry has benefited numerous Small to Medium-sized Enterprises (SMEs) around the world. However, there are several common barriers to SMEs to adopt e-commerce approach. The focus of the work is specifically on Physiological and Technological Barriers in the adoption of E-commerce within the British SMEs. Researching on this topic can deliver the direct knowledge base for the SMEs' owners to think over when making the relevant decisions.
- 1:40 An Analytical Study Evaluating the Applicability of a Developed Innovative E-Sourcing System for Automobile Based Firm
-
Over the years several studies have contributed to e-businesses, but at the moment there are fewer studies that have supporting e-sourcing operation in automobile based firms. Therefore, this paper aims to develop an innovative e-sourcing system for automobile based firms based on Model View Control (MVC) architecture to facilitate procurement operation among buyers and sales staffs. The developed system aims to lessen the work of sales staff thereby reducing tedious task faced by the sales staff. Additionally, the proposed system replaces the traditional system by reducing paperwork in the form of maintaining various files manually. The system also supports in maintaining and storing of customers' information. Besides, the system save time and facilitate customers in ordering automobile product whenever they want to without visiting the company. Moreover, the developed innovative e-sourcing system was evaluated using questionnaire data to assess the applicability of the system in managing automobile ordering between the customer and sales staff based on data collected from fifty respondents. Accordingly, descriptive analysis was carried out to analyze the collected data by deploying Microsoft Excel. Results from the analyzed data reveal that the developed innovative e-sourcing system is applicable in supporting sales of automobile based products.
- 2:00 Investigating the Impact of Contactless Payment Technologies on the Students' adoption at the University of Bahrain
-
The Kingdom of Bahrain witnessed many advancements in the FinTech domain recently. One of these obvious advancements is the introduction of few contactless payment software and technologies. Within the first two quarters of 2018, several institutions introduced contactless payment mobile apps such as BenefitPay, bWallet, VIVA Cash and MaxWallet. These mobile apps received vast adoption rate in a relatively short period of time by the citizens in the Kingdom. This study is conducted to investigate the impacts of contactless payment technologies on the students of University of Bahrain. The study focusses on investigating the adoption factors of contactless payment technologies and their impact on the willingness of students to adopt these technologies. For the operationalization of this study a modified version of Technology Acceptance Model (TAM) is proposed and tested through a survey, while the statistical package for social sciences (SPSS) is used for data analyses.
S4: Biomedical and Healthcare
Room: S45-11-Zain E-Learning Center
Chairs: Hessa Al-Junaid, Sid Ahmed Chouakri
- 1:00 A Mobile Cloud-based System for Alzheimer's Disease
-
The recent advances in cloud computing have been widely used in different domains and highly employed for designing cloud-based solutions in healthcare. With the could-based solutions, services can be accessed by patients from anywhere using their own mobiles. In this paper, a cloud-based system is presented as a solution to address the needs of Alzheimer's patients and their caregivers. Those needs are tackled by developing two mobile applications to streamline the care-giving process, and thus act as if the caregiver is in presence all the time with the patient. The privacy and confidentiality of this problem is also considered using the latest set of technologies available. The proposed solution as a cloud-based system has been proven to be practical, affordable, accurate and efficient.
- 1:20 The Influence of Content and Device Awareness on QoE for Medical Video Streaming over Small Cells: subjective and objective quality evaluations
-
Small cell networks are expected to be an integral part of future 5G networks in order to meet the increasingly high user demands for traffic volume, frequency efficiency, and energy and cost reductions. Small cell networks can play an important role in enhancing the Quality of Service (QoS) and Quality of Experience (QoE) in mobile health (m-health) applications, and in particular, in medical video streaming. This paper presents content-aware and device-aware medical Quality of Experience evaluations in terms of subjective (e.g. MOS) and objective (e.g. PSNR and SSIM) quality metrics obtained over small cell networks. Furthermore, we address the following two main research questions: (1) How significant is ultrasound video content type in determining medical QoE? (2) How much of a role does the display device play in medical experts' diagnostic experience? The former is answered through the content classification of ultrasound video sequences based on their spatio-temporal features and validating their significance through medical experts' subjective ratings. The latter is answered by conducting a subjective experiment of the ultrasound video sequences across multiple devices, ranging in screen size and resolution.
- 1:40 Psychological Stress Detection from Social Media Data using a Novel Hybrid Model
-
Psychological stress is considered as the biggest threat to individual's health. Hence, it is vital to detect and manage stress before it turns into severe problem. However, conventional stress detection strategies rely on psychological scales and physiological devices, which require active individual participation making it labor-consuming, complex and expensive. With the rapid growth of social networks, people are willing to share moods via social media platforms making it practicable to leverage online social interaction data for stress detection. The developed novel hybrid model Psychological Stress Detection (PSD), automatically detect the individual's psychological stress from social media. It comprises of three modules Probabilistic Naïve Bayes Classifier, Visual (Hue, Saturation, Value) and Social, to leverage text, image post and social interaction information we have defined the set of stress-related textual 'F = {f1, f2, f3, f4}', visual 'vF = {vf1, vf2}', social 'sf' to detect and predict stress from social media content. Experimental results show that the proposed PSD model improves the detection performance when compared to TensiStrength and Teenchat framework, PSD achieves 95% of Precision rate. PSD model would be useful in developing stress detection tools for mental health agencies and individuals
- 2:00 Proposed Disease Prediction System Based on Data Mining Algorithm
-
The development of Big data in biomedical and healthcare communities offers an exact analysis of medical data benefits such as primary disease detection, patient care, and public services. But, the analysis accuracy is reduced. The value of medical data is incomplete. Moreover, several regions exhibit unique features of certain regional diseases which may decline the prediction of disease outbreaks. In the proposed system, novel decision tree based algorithms are used which leads to considering more factors in general and predictions with high accuracy compared to other studies in Disease Prediction. It uses the 2 kinds of novel algorithms for classification which. Data mining are used for the medical applications like a prediction of disease, determination of disease treatment, and prediction of healthcare costs. Associated with other kinds of prediction, disease prediction is a significant method in today's medical world. The Disease Prediction has a substantial part in data mining. One of this research aims is to offer a novel KBayesTreeRegression based multimodal disease risk prediction (KBayesTreeRegression-MDRP) algorithm to provide the prediction accuracy of above 94.8% with a convergence speed, which is faster than existing techniques. . Recent technologies are utilized in the medical field to enhance the medical services in cost-effective manner. Data Mining methods are also utilized to inspect the numerous factors which are responsible for diseases for instance food type, diverse working environment, educations, living atmosphere, pure water availability, health care services, cultural, environmental and agricultural factors. Medical decision support systems are implemented to support clinicians in their diagnosis. They typically work through an examination of medical datasets and also knowledge base of clinical expertise. The excellence of medical diagnostic decisions can be augmented through improvements to these systems. Data mining provides a way to get the information suppressed in the data. They can find patterns hidden in large and complex collections of data, where these patterns elude traditional statistical approaches to analysis Keywords- Data Mining - Disease Prediction - Healthcare Community - System Performance
- 2:20 The Human Factor: Reconciling socio-cultural practices and technological innovations in m-Health Delivery for Underserved Communities
-
This study deals with the successful implementation of m-Health projects. Public health has become a cause for concern for many nations. Countries such as the United States and India among others have launched several initiatives in recent years with the aim to improve public health. However, these initiatives always face the problem of allocating scarce resources to a large number of patients. The study analyses 103 m-Health studies conducted between 2008 and 2016 across multiple databases. To maintain the accuracy and objectivity of results a text-mining approach is followed. The results of the analysis are then used to create questions for structured interviews which are then conducted with the different stakeholders of an m-Health project currently being operated in Ahmedabad, India. The daily operations of the system are also observed. The interviews and observations of the functioning of the system are then analyzed in Atlas.ti. The results of this analysis are then reconciled with the results of the text-mining to create a roadmap for m-Health delivery that avoids the current pitfalls that occur with m-Health projects.
Sunday, November 18 2:20 - 3:10
LB-1: Lunch Break Day-1
Room: S22-BTC Food Court
Sunday, November 18 3:10 - 4:50
S8: Internet of Things
Room: S45-108-Zain E-Learning Cente
Chairs: Mohamed Eltoweissy, Chitra Balakrishna
- 3:10 Design of Chipless RFID Tag Based on Stepped Impedance Resonator for IoT Applications
-
The current growth in the identification, following, and sensing applications for the Internet of Things (IoT) have pushed the chipless RFID technology to gain much attention due to its remarkable advantages compared to conventional barcodes. New chipless RFID tag based on steeped impedance resonator (SIR) is presented in this paper. This tag has a data capacity of 6-bits in the range from 2 to 2.6 GHz for a size of 5.4×2.6×0.05 cm3. The proposed tag consists of an SIR circuit and two cross-polarized ultra-wideband (UWB) antennas to transmit and receive the interrogation and response signal of the chipless Radio Frequency IDentification (RFID) tag. The prototype of the tag is designed on a Taconic TLX-8 substrate of dielectric constant 2.55 and thickness 0.5 mm. The proposed passive SIR chipless RFID tag can be used in many industrial and IoT applications.
- 3:30 A Proposal For Server-less Cloud-based Infrastructure For A Smart Metering System In The Kingdom Of Bahrain
-
Replacing the normal electricity and water meters with smart ones is one of the major transformations toward smart cities in the Kingdom of Bahrain. This transformation eliminates the need for manual reading and the estimated monthly billing values. The current approach of the system infrastructure is having the smart meters connected to a network and communicates with a server which reads the data from the meters on periodic times. The current system requires the connection to be active and the servers to be turned on all the time or at least for the time taken to connect to all meters or when any request is received from a user. Our proposed solution benefits from the server-less infrastructure which uses cloud managed services to eliminate the need for the server-based infrastructure. The proposed solution minimize the computing to its minimum requirement while enhancing the security for the system. In this paper, we are discussing the current system and the related issues with the required resources. Also, we are presenting the proposed system while comparing both systems for the used resources, security, complexity, availability, maintenance requirement, and reliability.
- 3:50 Zero-Trust Architecture for Internet of Things
-
Internet of Things (IoT) is growing exponentially. The world is starting to accept and adopt this new technology which promises better control over and information about our surroundings. However, many researchers are trying to grab attention towards the security risks associated. Most of the risks are due to the heterogeneity of IoT devices and the fact that most of them are computationally-constrained and power-constrained. This prevents them from using the same security standards as normal computers. Our research tackles the network aspect of IoT security. Current solutions do not account for internal threats, which necessitate the need to adopt more stringent security architectures. Our solution is to apply the novel zero-trust security (ZTS) model to IoT in order to protect from those threats. In this paper, we will study the risks related to the network security of IoT devices. We then present ZTS as a solution to mitigate those risks. A proof-of-concept architecture will be developed to depict how IoT devices will fit in within the ZTS context. Finally, a way forward for current IoT devices to shift to ZTS architecture will be laid out and discussed.
- 4:10 Homeland Security Video Surveillance System for Smart Cities
-
A novel Motion Estimation architecture, used for homeland security video surveillance systems, proposed in this paper. The proposed architecture is suitable to be adopted in H.265 High Efficient Video Coding (H.265/HEVC) encoder of video surveillance systems. The proposed architecture uses level A and level B data reuse to minimize memory I/O while fully utilizing the hardware resources. An efficient local memory is used for performing data reuse while loading both the current block and the search area inside the Processing Element (PE) array. The proposed architecture has a smooth and a regular data flow. The video resolution accuracy of the proposed architecture is very high. Subjective and quantitative measures are used to measure the performance of the proposed architecture. Our architecture can be adopted in the encoder of video surveillance systems and any High Quality (HQ) video related applications such as High Definition (HD) broadcasting. The time verification of the proposed architecture is tested using Modelism-version10.4a.
- 4:30 Internet of Things: Architectures, challenges and applications
-
Internet of Things (IoT) is one of the biggest and powerful technology inventions, it is basically a number of networks that connect the devices, vehicles, buildings and other aspects of our lives. This could be in the form of sensors, software, tags and network connectivity. As we are living in a glorious and prosperous era and moving towards a big evolution in life, where the world is being governed, ruled, and manipulated by the internet forces and the new computing inventions. This paper presents a comprehensive overview of IoT and the main enabling technologies. Also, it describes the IoT architecture, its security concerns and mentions the related key challenges of the heterogeneous services that provided by IoT enabled assisted living technologies specifically in smart homes.
S7: Cloud Computing
Room: S45-06-Zain E-Learning Center
Chairs: Muain Aljamlan, Fawzi Mohammed Munir Al-Naima
- 3:10 Is Bitcoin Mining Halal? Investigating the Sharia Compliance of Bitcoin Mining
-
Cloud Computing has powered a wide range disruptive technology that changed the economic landscape of many industries. One of these disruptive technologies is Block Chains and the emergence of cryptocurrencies like Bitcoins. This research explores the Islamic perspective of Bitcoin Mining. For a Bitcoin transaction to be processed it needs to be 'mined', which is the process to solving a mathematical equation as well as approve Bitcoin transactions distributing it over various ledgers. The research will highlight the notion of anonymity of the source and destination of the funds in Bitcoin which seen as problematic with Islamic Banking scholars. Especially since anonymity is the cornerstone of Bitcoin and the feature that allowed Bitcoin and other cryptocurrencies to spread. This research will rely on leading Islamic Banking scholars in the Kingdom of Bahrain to evaluate Bitcoin Mining's compliance with Sharia.
- 3:30 Performance Evaluation of Datacenter Network Topologies via NS-2 Simulations
-
A datacenter provides the core computing and storage elements for cloud computing paradigm. The Datacenter network (DCN) topology defines the structure in which servers and the networking devices are interconnected within a data center. This paper presents the results of a comparative simulation study for four well-known DCN topologies, basic tree, Google fat tree, Facebook fat tree and Dcell, using a NS2 simulation environment. The traffic patterns used in the simulations are one-to-one, one-to-all and the all-to-all. The simulation results for these four topologies are also compared under various parameters changes including packet size and the TCP window size. The simulation results capture the performance metrics which are the average packet delay and the throughput. The simulation study shows that for less than 20 servers, the Dcell topology has smaller latency and higher throughput compared to the other topologies, while the Facebook fat tree topology performs better when the number of servers in the data center is large.
- 3:50 Friendly Online Technology Development Cloud Service for Bahraini Students based on E-Advisor
-
Cloud Computing is a modern technology that spread over the world through various Cloud services. Everything as a service (Xaas) is the magic of Cloud Computing. Today, many of the E-Accessibility is running with help of Cloud Computing's Resource Management Technique. Resource Management Technique has resource selection, resource matchmaking, resource monitoring, resource scheduling, resource allocation and resource accessibility. The proposed Technology Development service is working based on Elastic Resource Management Technique, E-Advisor and priority based Mapreduce. The students are very difficult to reach the Technology Development services in worldwide. This work proposes a unique Cloud portal for Technology Development services for Bahraini Students based on E-Advisor. Here E- Advisor is act as a Cloud broker and it's controlled by the government. This simulated work is implemented in Cloudsim with proposed priority based Mapreduce algorithms and increased Quality of Services (QoS) Attributes. This online Cloud service is very useful for Bahraini Students, staff and Industry experts. It produces the high impact in the education field in Bahrain with title of 'Bahrain Technology Development 2020'. This Cloud portal provides an enhanced and accessible service for students.
- 4:10 How to Improve the Resource Utilization in Cloud Data Center?
-
Nowadays all departmental applications are migrating to the Cloud services due to the more efficiency. Cloud utilizes more number of data centers for the storage, computation and service. Data center consists of various Cloud Resources to achieve the tasks. Resource Utilization is an important Consideration to save the energy of resources in Data Center. The Cloud Computing has different Techniques to utilize the resources efficiently. This research work use to improve the resource utilization factor with help of Quality of Service (QoS) parameters. It also saves the time and energy of resources in the Green Cloud Computing network. Resource providers should follow the Green Service Level Agreement (GSLA) in this proposed Cloud Network. This proposed research work is follows the Green Resource Management Techniques (GRMT). Green Resource Management Techniques (GRMT) is consists of green based adaptive fault tolerance resource selection, green based virtualization of resource allocation, green based match making and scheduling algorithms, green based cloud balancing, green based cloud technology. This work handle only two components, such as green based adaptive fault tolerance resource selection and green based virtualization of resource allocation. It is used to decreases the energy consumption, Carbon dioxide (CO2) emission and dynamic cost.
- 4:30 Challenges of Large Scale Data Processing in the Cloud
-
We are living in a world of big data. The word "big data" not only means a huge amount of data but also their representation and the velocity of access also differs from that of traditional data. Many different approaches have been proposed for data distribution, management, and processing of large-scale data processing applications. The large-scale data processing applications can be implemented in in-house clusters or in the cloud. Cloud computing offers large-scale data processing services in an economic way. This paper gives an insight into these approaches and their challenges.
Monday, November 19
Monday, November 19 8:00 - 8:30
R2: Registration-Day2
Room: S45-Zain E-Learning Center
Monday, November 19 8:30 - 8:50
KS-4: Keynote Speaker
Artificial Intelligence Research at the Communication and Information Research Center
Prof. Abdulnasir Hossen, Sultan Qaboos University, Sultanate of Oman
Room: S45-101-Zain E-Learning CenterChair: Abdulla Alqaddoumi
The Communication and Information Research Center (CIRC) at Sultan Qaboos University (SQU) in Oman was established in 2002 to promote and advance research and education through Government/University/ Industry partnerships in focused and shared competitive ICT research programs. A number of challenges on the horizon are FOSS deployment, IoT, smart cities, Artificial Intelligence, telemedicine, mobile multimedia access, quality of service, information security. The presentation will cover the current and previous research work and community service activities at CIRC, besides the new plans to collaborate with other institutes in the region. Emphasis on certain research activities in artificial intelligence will be covered during the talk. What is AI, new definitions, what is the AI Industry revolution 4.0, Automation, what are the necessary future skillsets and areas of investments. New technologies and labor market. Areas of AI studies at SQU. Proposed AI research in Education. Artificial intelligence in medicine and use of artificial neural networks in diagnosis. Non-invasive diagnosis techniques using biomedical signal processing are required in hospitals not only to simplify diagnosis methods and to reduce pressure on hospitals, but also to have more accurate automatic systems for diagnosis. Examples on identification and classifications of patients with obstructive sleep apnea, congestive heart failure, preeclampsia, parkinsonian tremors will be given in the presentation.
Monday, November 19 8:55 - 9:20
KS-5: Keynote Speaker
IoT and Cloud based open source platforms for healthy and independent living - Present and Future
Dr Princy Johnson is a Senior Lecturer in the School of Engineering at Liverpool John Moores University, UK
Room: S45-101-Zain E-Learning CenterChair: Aisha Bushager
The provision of flexible platforms to support smart and healthy living at home is increasingly important both to improve quality of life for citizens and to ensure that they remain productive when working and to enable the development of sustainable health and social care systems. Personal pressures arise as individuals age and begin to experience age-related conditions, often concurrently. Some of these conditions are caused by or accelerated by lifestyles and behaviour but the impact can, in many cases, be mitigated by effecting changes in lifestyle and behaviour choices. Currently, there are several commercial and government funded projects to enable development of open source flexible platforms. The search is still not over. What novel processes and techniques are needed to aid in the development of such large scale platforms. This keynote will touch upon innovative techniques and methodologies that will support development of such scalable learning platforms
Monday, November 19 9:20 - 9:45
SB-3: Short Break
Room: S45-Zain E-Learning Center
Monday, November 19 9:45 - 11:25
S9: Cyber Security
Room: S45-108-Zain E-Learning Cente
Chairs: Hesham al-Ammal, Mazen Ali
- 9:45 Using Keystroke Authentication Typing Errors Pattern as Non-Repudiation in Computing Forensics
-
Access to information and data is becoming an essential part of nearly every aspect of modern business operations. Unfortunately, accessing information systems comes with increased chances of intrusion and unauthorized access. Acquiring and maintaining evidence from a computer or networks in the current high-tech world is essential in any comprehensive forensic investigation. Software and hardware tools are used to easily manage the evidence and view all relevant files. In an effort to enhance computer access security, keystroke authentication, is one of the biometric solutions that were proposed as a solution for enhancing users' identification. This research proposes using user's keystroke errors to determine guilt during forensics investigations, where it was found that individuals keystroke patters are repeatable and variant from those of others, and that keystroke patterns are impossible to steal or imitate. So, in this paper, we investigate the effectiveness of relying on "user's mistakes" as another behavioral biometric keystroke dynamic.
- 10:10 The role of User Entity Behavior Analytics to detect network attacks in real time
-
This paper aims to assess the value and success of using behavior analytics in securing the network from not-before-seen attacks such as zero-day attacks. This paper uses a systematic literature review and self-administrated survey and interviews with convenience sampling of high profile network users and top security vendors. Survey and interviews with various security experts are utilized to verify the matter-of-fact effectiveness of the solutions based on behavior analytics. During collecting the primary data via a survey, researchers will go for a structured interview with vendors who are selling solutions to understand the performance of behavior analytics-based solutions and the distinct features of their solutions.The endeavor of this paper is to highlight the weaknesses and strengths of different UEBA solutions and their effectiveness for detecting network attacks in real-time interaction. This research contrasts top fifteen UEBA technologies based on use cases and capabilities and highlights common usage scenarios. Based on the evidence, recommendations will be given.
- 10:35 Multiple Encrypted Random Forests using Compressed Sensing for Private Classification
-
Recently there has been a huge interest in private computing. In particular; to perform privacy preserving classification tasks. Homomorphic encryption offers secure asymmetric encryption solution to this problem, however, it comes with a high storage and computation cost. Compressed sensing (CS) on the other hand is much lighter; however, lacks privacy since the encryption uses a symmetric key which is the random projection matrix. A novel privacy preserving classification approach is proposed in this paper which employs multiple encrypted random forests with two compressed sensing encryption levels. At the first level, each feature vector is CS-encrypted using a different random matrix for each forest. At query time, the user selects one matrix randomly from a set of R different matrices, and gets R encrypted results of which only one will be used. At the second level, the class-label information at each tree leaf is encrypted using a different CS matrix for each tree. During recognition, the cloud adds all the encrypted leaf CS vectors for each of the R forests and sends to the user, where only one of them goes through sparse recovery to find the class label. Experiments on COREL1K and CIFAR10 image classification tasks show that the proposed approach achieves classification accuracy similar or better than nearest neighbour classifier with plaintext features. Also, correlation-based results show that the multiple-forest approach offers good level of semantic security between the feature vectors while the class-label CS encryption approach offers excellent privacy to prevent the cloud from knowing the class-label.
- 11:00 The influencing factors of internet banking in dubai
-
In this paper, the effectiveness of human factors on the e-banking system in Dubai had been identified and analyzed. The study targets are to aspect the human factors by literature review. Moreover, the statistical analysis has been studied for these factors. The study has focused on the costumers have dealing with the local and international banks in Dubai and utilized survey of questionnaire to get the significant information. The investigation to uncover security mindfulness among the clients of e-banking in Dubai and to discover out the effect of the components on the choice of the clients to utilize the electronic framework.
S10: Telecommunication and Networks-2
Room: S45-06-Zain E-Learning Center
Chairs: Magdy Bayoumi, Akram Zeki
- 9:45 5G and Effect of Key Coexistence Factors
-
The main goal of this paper is to address the spectrum coexistence issue of 5G systems by investigating the intersystem interference phenomenon. The intersystem interference is highly highlighted in two considered frequency bands, 800- and 3500-MHz, due to the congested various wireless systems within these spectrum bands in addition that the 5G systems will employ all available frequency spectrum including the millimeter waves. Some main coexistence factors will be analyzed and interpreted in related to the coexistence scenarios between 5G and a wireless fixed link. Results are provided for the sake of more understanding and comprehension of the coexistence issue.
- 10:05 X-band Conformal Antenna Array For Low Cost CubeSats
-
This paper present, a design of X-band conformal slot antenna array for cube satellite. To increase the perfect attachment of the antenna onto the platform of unit two (U2) CubeSat, the low profile broadband feeding structure is used to reduce EMC issues. The array of 1×6 elements has been simulated in HFSS with the gain of 11.15dB and SLL of 19.62dB. the center frequency is 10 GHz and VSWR is ≤2
- 10:25 Quantized Run Length Encoding QRLE -New Compression Method-: Application to ECG Transmission via IEEE802.11b WLAN Channel
-
A new method of signal compression, called Quantized Run Length Encoding QRLE, based upon the 'classical' run length method combined to discrete wavelet transform thresholding, intended for a simulated transmission via the IEEE 802.11b WLAN channel, is presented in this work. The key idea of our new method consists of quantifying each pair of zero followed by its corresponding run number, issued from the RLE, by one value consisting on the run number value plus an offset of predetermined integer value. In this work, the suitable offset was adjusted to 1024(≡210). This leads to quantizing the 'new' run number on 11 bits while the non null ECG thresholded ECG signal samples are quantized on 10 bits. The trivial advantage of this method is suppression, and consequently gaining, of all zeros. We have applied the algorithm to real ECG signals, exhibiting some cardiac status, extracted from the MIT-BIH arrhythmia data base, transmitted via a 'simulated' IEEE 802.11b WLAN channel. To evaluate the new QRLE based ECG compression algorithm, we have simulated transmitting the compressed ECG signal via the IEEE 802.11b WLAN channel. In terms of compression efficiency, the algorithm achieves compression ratio of around 10:1, normalized root mean squared error (NRMSE) of 0.1% and (mean± standard deviation) of the difference between the restituted ECG signal and the original one of around (10-4) ± 0.02. Moreover, a comparative study with respect to selected ECG compression algorithms show the higher performance of the developed new technique called 'Quantized run length encoding QRLE'.
- 10:45 Road Lane-Lines Detection in Real-Time for Advanced Driving Assistance Systems
-
In this paper, a fast and reliable lane-lines detection and tracking technique is proposed. The proposed technique is well suited to be used in Advanced Driving Assistance Systems (ADAS) or self-driving cars. The main emphasis of the proposed technique in on simplicity and fast computation capability so that it can be embedded in affordable CPUs that are employed by ADAS systems. The proposed technique is mainly a pipeline of computer vision algorithms that augment each other and take in raw RGB images to produce the required lane-line segments that represent the boundary of the road for the car. Each used algorithm is described in details, implemented and its performance is evaluated using actual road images and videos captured by the front mounted camera of the car. The whole pipeline performance is also tested and evaluated on real videos. The evaluation of the proposed technique shows that it reliably detects and tracks road boundaries under various conditions. The usefulness and the shortcomings of the proposed technique are also discussed in details.
- 11:05 Comparison of 2D and 3D Propagation in Wi-Fi Networks
-
Accurate and realistic modelling of the wireless channel is vital for precise evaluation of the system level performance of wireless transmission systems. Therefore, channel models have evolved from basic theoretic empirical models that focused on large scale fading to more advanced and standardized models. In addition, the introduction of advanced Multiple-Input Multiple Output (MIMO) techniques such as massive MIMO, Full Dimension (FD)-MIMO, and 3D beamforming requires the development of accurate 3D wireless channel models to fully exploits these MIMO technologies. This paper highlights the impact of 3D propagation on MIMO performance in Wi-Fi system, precisely in terms of physical (PHY) throughout and Signal-to-Noise Ratio (SNR).
S11: Informatics-2 (Education & E-Learning Systems)
Room: S45-11-Zain E-Learning Center
Chairs: Mohammed Redha Qader, Athraa Almosawi
- 9:45 Integration of ethics in e-Learning through Virtual Academic Counsellor
-
It is vital for e-Learning stakeholders to overcome the factors which influences the students to deviate from educational objective. A Delphi study with the help of e-Learning experts was conducted to investigate and control this issue. It was found that there is deficiency of face-to-face interactions among students and teachers in e-Learning. Moreover, lack of counselling and traditional policies in independent e-learning environment also become the cause of this ethical dilemma. Different approaches to control this issue were also discussed and investigated to propose a model. It is recommended to involve members of the society as virtual academic counsellor in e-Learning to develop ethical and moral values of students. The study fulfils an identified need that how ethics can be integrated in e-Learning management system of academic institution to produce ethically competent graduates. It is expected that by following provided guidelines, the passing graduates will have field specific knowledge along with good ethical building to serve not just their families and societies but the entire nation.
- 10:05 Evaluating Students' Satisfaction of E-Learning Using Decision Support Systems (DSS)
-
E-learning is a form of education that is increasingly being used in higher education in the developed world. The aim of the study was to evaluating students' satisfaction of e-Learning. In this research, we apply and use the theory of technology acceptance model (TAM). We employ structural equation modelling (SEM) approach with SmartPLS software to investigate students' adoption process. Findings indicates that the perceived ease of use, perceived usefulness and intention to use e-learning among university students have a positive impact and substantially associated with learning performance and learning satisfaction. The study concludes that university students in Malaysia have positive perceptions towards e-learning and intend to practice it for educational purposes.
- 10:25 A Multifunction Reminder Platform for an Educational Environment
-
In this paper, I present a multifunction reminder platform that is dedicated to an educational environment. I consider the reminder as a two-dimensional function, i.e. depending on two variables, time and space. It automatically produces alarms with notifications whenever these two variables take specific values. In other words, when a student approaches an educational or administrative location based on his GPS geo coordinates. The two-dimensional function reminder helps a student to better manage time and avoid delays in accomplishing tasks. I identified different educational uses for such a reminder that relate to the daily life of a student. The reminder function is demonstrated for the library case study.
- 10:45 Accuracy vs. Cost in Decision Trees: A Survey
-
Decision Trees have been applied widely for classification in many fields such as finance, marketing, engineering and medicine. The increased field of application, made the requirement for understanding various aspects of decision trees in deep. This paper introduces the concept of decision trees, their various areas of application in data mining, summarize the standard decision tree algorithms, and identify their main advantages and disadvantages. It mainly aimed to clarify relationship between the classification accuracy and classification cost in decision trees as balancing the two is very important in many fields such as medical diagnosis.
- 11:05 Hybrid Circuit Simulation and Modeling Cloud-Based System Using an integrated IoT for Electronics and Computer Engineering e-Learning Courses
-
This paper shows the design and implementation of hybrid circuits cloud-based enables students to design, simulate, and model both Analog and Digital circuits and integrates the IoT by adding ability to include sensors and interact with the design on real laboratory to act as a comprehensive laboratory device to draw circuits, integrate different components, apply different inputs and simulate, capture, measure, and detect any defect or heated elements. The system scales by elastic features of the cloud computing. The cloud system proposes an integrated environment for digital, analog and internet-of-things modeling, integration and decomposition. The cloud system to show how cloud-based simulator can transform the eLearning and enhance the digital learning teaching and studying environments. The lab device can be integrated into circuits level with the cloud system to run real circuits simultaneously with the simulation. provide in-browser cloud system for schematic capture and circuit simulation & emulation. As a testimony of the system of the modeling and simulation discipline, a list of about 100 types of equipment's included and tested.
Monday, November 19 9:45 - 4:00
S12: Internet of Things - Workshop
Room: S45-12-Zain E-Learning Center
Chair: Chitra Balakrishna
- 9:45 International Workshop on Internet of Things (IoT): Technology and Applications
-
An understanding of the way the IoT systems work is necessary for end users to enable them to maximise their benefits from IoT. This workshop on IoT serves as a boot strap to initiate end users into the IoT ecosystem and appreciate its utility, characteristics as well as risks. The workshops are targeted towards industry professionals, faculty, students, and enthusiasts. The workshop content is designed with the current market trends and provides information on use-case scenarios and corresponding solutions. The workshop content provides sufficient awareness to the participants to be able to engage in discussions for implementation of functional IoT solutions. The workshop addresses the generic approach to building an IoT system. It is intended to help users to understand solutions for specific problem areas in smart home, industrial automation, transport and mobility, healthcare and in general, smart city initiatives. The workshop will include a series of talks and a couple of Q&A sessions.
- 11:25 P2 Prayer
Monday, November 19 11:25 - 11:50
P2: Prayer
Room: S45-Zain E-Learning Center
Monday, November 19 11:50 - 4:30
PS-2: Poster Session
Room: S45-Zain E-Learning Center
Monday, November 19 11:50 - 1:10
S13: Machine Learning and Robotics
Room: S45-108-Zain E-Learning Cente
Chairs: Nabil Hewahi, Ebrahim Abdulla Mattar
- 11:50 Performance Analysis of Machine Learning Classifiers for ASD Screening using Rapidminer
-
Several machine learning classifiers have been used for Autism Spectrum Disorder screening, however, literature in finding the best classifier for this application domain is limited. Hence, this paper presents a comparison of five (5) supervised machine learning algorithms: Decision Tree, Naïve Bayes, k-nn, Random Tree, and Deep Learning using small datasets (n=1100) on child, adolescent and adult ASD screening in finding the most appropriate classifier. These algorithms, which are evaluated using a broad set of prediction performance metrics including accuracy, precision/recall measures, and Receiver Operating Characteristics, are compared against each other. The experiment result suggests that the Deep Learning classifier gives the best performance (with more than 96%) in almost all metrics while the Random Tree classifier came out as the least performing classifier in all the performance metrics.
- 12:10 Detection of Arabic Spam Tweets Using Word Embedding and Machine Learning
-
In this study, we explore the idea of applying word embedding based features with machine learning techniques to detect Arabic spam tweets. In addition, the effects of text domain of the collected corpus to learn word embedding techniques is analyzed. This is evaluated using a publicly available dataset of 3503 tweets alongside with three popular classifiers for binary classification problem, namely: Na\"ive Bayes, Decision tress and SVM. The experimental results revel that word embeddings based features able to recognize tweets generated automatically from those generated by human and outperform the baseline. An accuracy rate of 87.33% is achieved using skip-gram word2vec technique with SVM. In addition, word embedding models are domain independent such that the models learnt using corpus collected from twitter outperform others significantly.
- 12:30 An Optimized Machine Learning Approach using Kohonen's Self Organizing Map showing Gene Dependency for Cancer Mediation Biomarkers
-
All Cancers are genetic, which means that they result from the unnatural function of one or more genes. For thousands of years, lots of effort has been employed by a number of researchers to identify the possible set of biomarkers for cancer detection. To analyse such a vast set of gene expression, a number of machine learning techniques are being used. In this article, we have employed Kohonen's Self Organising Map mechanism to optimize the huge dataset of gene expressions to provide an easy way of representing multidimensional dataset in much lower dimensional spaces. Furthermore, we also have demonstrated an important test to determine whether a test sample of genes is cancerous or not. By mapping the test sample with a few cancer affected gene samples, we have clustered the optimized data set to check the number of clusters, which has helped us identifying the genes which are actually affected by cancer disease. Hence, this paper reports an easy optimization technique to identify the cancer affected genes thereby, facilitating easy diagnosis of cancer.
- 12:50 Single Machine Scheduling Problem in Uncertain Environment
-
The most current models of the combinatorial optimisation problems assume that the input data is perfectly known and reliable. In practice, this assumption is often wrong since only data estimations are available or the data is often subject to future changes due to some online disturbances. In this paper, we consider the scheduling problem of minimising the number of tardy tasks on a single machine with uncertain data. The uncertainty is related to an unavailability of the machine which can happen at the beginning of the schedule. We will discuss the robustness approach and we will present different robust solutions with different criteria for the studied problem
S14: Optimization, Modeling and Simulation
Room: S45-06-Zain E-Learning Center
Chairs: Alauddin Yousif Al-Omary, Youssef Harrath
- 11:50 A Novel Graph-Based Algorithm for Solving the Right Partner Problem
-
This research work investigates the problem of selecting the right partner problem. It consists of pairing $n$ given students based on their choices of likes and dislikes with the aim to maximizing the degree of overall satisfaction of students' choices. The problem is modeled using graphs with special weights on edges and vertices. Two graph-based algorithms are proposed and tested using various problem instances of different sizes. The experimental results showed very promising performances for both algorithms.
- 12:10 A solution to skewed load distribution problem in shortest path methods
-
Ad hoc networks are infrastructureless and lack any central administration. Due to scarcity of resources, most tasks require distributed participation. The shortest path methods are the most widely used routing methods due to their simplicity and efficient operation under low traffic conditions. But as the load in the network increases, it puts stress on the few nodes which form the shortest path. Most of these overloaded nodes tend to lie towards the centre of network. This creates a skewed load distribution in such networks and affects the overall performance. In this paper, we confirm the skewed load behaviour of one such shortest path method for ad hoc networks. We analyze the load on different nodes and relate them to their position in the network. Our experiment verifies a strong negative correlation between distance of node from network centre and its RouteCache table size. We also propose a metric called RouteCache Index (RCI), which estimates the load (connections) on a node. Then we propose a multipath routing method, which uses the RCI metric. This proposed method overcomes the skewed load behaviour and improves the performance of such networks.
- 12:30 QCI and QoS Aware Downlink Packet Scheduling Algorithms for Multi-Traffic Classes over 4G and beyond Wireless Networks
-
The recent advancements in wireless technologies and applications make downlink scheduling and resource allocation a hot topic of research. Hence, fair scheduling and balanced Quality of Service (QoS) delivery for different types of traffic (e.g., video, VoIP, and best-effort) are vital for the next generation wireless networks. In this paper, we compare and analyse different downlink scheduling strategies in terms of network-centric performance metrics such as average packet loss ratio, average throughput, system fairness, and system spectral efficiency. In addition, we show the effect of the QoS Class Identifier (QCI) parameters on different delay-aware scheduling algorithms. Furthermore, we propose to modify the Log-rule, Linear-rule, and Modified Largest Weighted Delay First (MLWDF) scheduling strategies by including the QCI parameters in order to balance the QoS delivery for different traffic-classes with improvement to the overall system performance. Through simulation, we show that the proposed scheduling algorithms utilising the QCI and QoS parameters introduce a remarkable multi-objective improvement of the QoS performance parameters for different traffic classes (i.e. real-time and non-real-time).
- 12:50 The Simulation Models for Human Pedestrian Movement of a Departure Process in an Airport
-
The emergence of simulating human movement and behaviour rose in order to imitate and replace human interference with this complex demeanour. It is so diverse and uncontrollable to the extent it is almost impossible to undergo exhaustive experiments. Hence, simulating this model is crucial. Especially in designing a public area where human density is too much concentration and the need for properly planned traffic or human movement is very much needed. For that, simulations and modelling are always the best options for investigating human behaviour. In this research work, the aim is to investigate how effective and to what extent the ABS, DES and hybrid of both give benefits to the society adequately. Its expected results will be an improved and better version of human pedestrian movement in the airport of the departure process simulation model.
S15: Future Vehicular Communications & Self Driving Cars
Room: S45-11-Zain E-Learning Center
Chairs: Sarah Al-Shareeda, Etienne Alain Feukeu
- 11:50 Broadcast Storm Mitigation in VANETs Using RMS
-
The necessity to improve safety measures and reduce accident risks on public roads has prompted and identified the Wireless Access in Vehicular Environment (WAVE) technology as a prominent and efficient candidate for future vehicular communication. In view to achieve its main objective under WAVE standard, all nodes or mobile involved should exchange or broadcast safety messages at the rate of l to l0 Hz based on the congestion state of the network. However, in a congested environment, the Broadcast Storm (BSt) problem naturally occurs and exhausts the available channel bandwidth thereby rendering exchange and sharing of vital information impossible. In our recent work, the Dynamic Broadcast Storm Mitigation Algorithm (DBSMA) proved to be a good candidate for the BSt problem mitigation for Cooperative Awareness Messages (CAMs) only. The present work improved the DBSMA (IDBSMA) to cater for both CAMs and DENMs type messages. The developed IDBSMA approach was tested under two cases scenario conditions and simulation results demonstrated the outperformance of the IDBSMA over the DBSMA algorithm in three dimensions. Firstly, in terms of broadcasting delay, secondly in terms of performance within the preset range and lastly in terms of Reserved Margin Space (RMS) to account for the DENMs messages.
- 12:10 The Simulation on Vehicular Traffic Congestion Using Discrete Event Simulation (DES): A Case Study
-
The bottleneck networks of the vehicular traffic flow required the clear understanding and the insights of congestion factors, determines the time and location of traffic breakdown. The congestion propagates through the network caused by the numbers of vehicles continuously growing. There have been lots of research studies to resolve the issues. Simulation and modelling to the related issues seem significance to the operational research specifically the traffic flow dynamic for the design, analysis and the management. Hence, this work focused on a simulation model for the vehicular traffic flow in order to reduce the traffic congestion for a smooth and low-density traffic flow system. Three scenarios from a traffic flow system based on queuing environment which is current traffic flow system as S1, the newly proposed models as S2 and S3 have been suggested to improve the traffic flow of the intersection. The method of comparing the performance of simulation results and difficulties, alternative road, the predictive strategy as decision support model with the potential alternative road has been identified with developing traffic flow simulation using Discrete Event Simulation (DES). DES is as the well-known traditional approach to catering the problems regarding the queuing system, and run simulation models on the different traffic scenarios for bottleneck analysis. The lower values for the cumulative average all the parameters indicated the traffic congestion is decreased and to avoid the traffic jam. The experimental results and statistical analysis showed that the proposed simulation model S2 and S3 for vehicular traffic flow is better than the current traffic flow system S1. It can be summarized that model 3 (S3) can be applied to the new vehicular traffic flow system with the better simulation results has been produced from the experimental simulations.
- 12:30 Behavior Cloning for Autonomous Driving using Convolutional Neural Networks
-
In this paper, we propose using a Convolutional Neural Network (CNN) to learn safe driving behavior and smooth steering maneuvering as an empowerment of autonomous driving technologies. The training data is collected from a front-facing camera and the steering commands issued by an experienced driver driving in traffic as well as urban roads. This data is then used to train the proposed CNN to facilitate what we call it behavioral cloning. The proposed Behavior Cloning CNN is named as "BCNet" and its deep seventeen-layer architecture has been selected after extensive trials. The BCNet got trained using Adam's optimization algorithm as a variant of the Scholastic Gradient Descent (SGD) technique. The paper goes through the development and training process in details and shows the image processing pipeline harnessed in the development. The proposed approach proved successful in cloning the driving behavior embedded in the training data set after extensive simulations.
- 12:50 Tuning of PID Track Followers for Autonomous Driving
-
In this paper, a Proportional-Integral-Differential (PID) controller that facilitates track maneuvering for self-driving cars is proposed. Three different design approaches are used to find and tune the controller hyper-parameters, one of them is specifically proposed in this paper for this specific application. The proposed controller uses only the Cross-Track-Error (CTE) as an input to the controller, whereas the output is the steering command. Extensive simulation studies in complex tracks with many sharp turns, have been carried out to evaluate the performance of the proposed controller at different speeds. The analysis shows that the proposed technique outperforms the other ones. The usefulness and the shortcomings of the proposed tuning mechanism are also discussed in details.
Monday, November 19 1:10 - 2:00
LB-2: Lunch Break Day-2
Room: S22-BTC Food Court
Monday, November 19 2:00 - 3:20
S16: Software Engineering and Cloud Computing
Room: S45-108-Zain E-Learning Cente
Chairs: Faisal Alqaed, Aisha Bushager
- 2:00 Bahrain Parks Cloud Platform: Live Suit of Tools for Planning, Designing, and Managing Public Parks
-
The paper introduces an open source cloud platform for the efficient, knowledge-based, sustainable and inclusive development of urban parks - Bahrain Parks developed at the University of Bahrain. We describe the local context at the genesis of the initiative, its components, the rationale for the creation of a comprehensive, live, database, and the suite of web applications that deliver information to different audiences from professionals to the general public. We describe the data structure on MySQL hosting and the development of both backend and frontend in R Shiny, and the various web applications and how these leverage the data to useful information for supporting decision-making on park maintenance, planning, and design. We describe the technologies used. We share findings regarding the official acceptance and attempts to integrate the application in workflows at several agencies. We argue that the project belongs to the sphere of e-governance. To our knowledge, the project is unique as it is being developed to incorporate an evidence-based design process, comprehensively. Furthermore, it embeds explicit and implicit public participation in decision-making, through park rating and usage.
- 2:20 Can Blockchain Spark off the Reincarnation of India's Living Dead?
-
A large number of studies have found a negative correlation between economic growth and corruption. Therefore, governments implement various anti-corruption measures particularly technology-based mechanisms. The government of India has launched various initiatives such as Digital India and Digital India Land Records Modernization Program (DILRMP) to modernize land records in the country and curb the corruption at various levels of land transactions. However, despite a push for reform through the DILRMP, India's current land title management system remains plagued with deficiencies. The current system of land records and property ownership is rife with corruption. DILRMP is heavily dependent on the government functionaries to act as a trusted third party for the verification of data and processing of deeds. This leaves a space for unscrupulous elements to trick the system and register a fraudulent deed, transfer land titles, and so on. This paper investigates the usefulness of an emerging technology called Blockchain and its integration with India's biometric identity program called Aadhaar for management of land records and property registration. This paper proceeds with relevant case studies, by identifying the requirements and outlining the concept and system architecture for such a solution. From our analysis, we conclude that the blockchain technology in association with Aadhaar would be an effective mechanism to bring in much-needed transparency in the government functionaries, eradicate fraud and corruption, and enhance socioeconomic benefits.
- 2:40 Resource Management in Internet of Things: A Routing Perspective
-
Internet of Things (IoT), the latest and most popular development in the field of Information and Communication Technology (ICT), requires the intermingling of different technologies like networking, cloud, and analytics. Networking portion of IoT is handled mainly by the Low Power Lossy Networks (LLNs), which in turn plays notable role in the proper functioning of other components including cloud platforms. As a result, resource management within LLNs is an issue that can't afford to be left out. The paper analyzes this concern through an elaborate study of LLN routing protocols including RPL, LOADng, and opportunistic routing protocols. The study has been conducted so as to flash light on resource management capabilities of these protocols. Accordingly, a mapping has been provided between the features of different protocols and the resources managed by each of those features. Finally, the paper enlists concerns demanding immediate attention, coupled with possible solutions, so as to ensure the efficient functioning of an approaching smart world.
- 3:00 Locating Bug IDs and Development Logs in Open Source Software Projects (OSS): An Experience Report
-
The development logs of software projects, contained in Version Control (VC) systems can be severely incomplete when tracking bugs, especially in open source projects, resulting in a reduced traceability of defects. Other times, such logs can contain bug information that is not available in bug tracking system (BT system) repositories, and vice-versa: if development logs and BT system data were used together, researchers and practitioners often would have a larger set of bug IDs for a software project, and a better picture of a bug life cycle, its evolution and maintenance. Considering a sample of 10 OSS projects and their development logs and BT systems data, thus the two objectives of this paper are (i) to determine which of the keywords 'Fix', 'Bug' or the '#' identifier provide the better precision; and (ii) to analyse their respective precision and recall at locating the larger amount possible of bug IDs manually. Overall, our results suggest that the use of the '#' identifier in conjunction with the bug ID digits (e.g., #1234) is more precise for locating bugs in development logs, than the use of the 'Bug' and 'Fix' keywords. Such keywords are indeed present in the development logs, but they are less useful when trying to connect the development actions with the bug traces in software project.
S17: Data Mining and Knowledge Representation
Room: S45-06-Zain E-Learning Center
Chairs: Ahmed M. Zeki, Hadeel Alobaidy
- 2:00 Data Mining for Intelligent Academic Advising from Noisy Dataset
-
This paper proposes an approach to address the common problem of classification from educational dataset with irrelevant or redundant attributes. In particular, the paper focuses on using this approach to improve academic advising through intelligent prediction of students' performance. This is achieved by using a composition of the two popular classification methods: Decision Trees and Naïve Bayesian Classifiers. Naïve Bayesian Classifier is built using the significant attributes in the students' dataset that are identified by the Decision Tree. The results show that the proposed approach succeeded in identifying the main attributes affecting students' performance. Moreover, the evaluation results show that using a composition of Naïve Bayesian and Decision Trees in one approach outperformed the results of each classifier individually.
- 2:20 Ontology Driven Support to Association Rules
-
One of the data mining techniques - to find associative combinations of items - is called Association Rules. For this problem, there are several algorithms such as Apriori, FP - Growth, and CT-Pro. One of the advantages of the Apriori algorithm is that it produces maximum rules. Ontology is a set of hierarchically structured terms to describe a domain that can be used as the basic framework of a knowledge base. In this work, we illustrate that how the hierarchical type of ontology can be utilized by the Apriori algorithm to improve the results. The Apriori with ontology implements the IR which is a parameter to determine the degree of association between combinations of items in a dataset. The series of experiments show that the proposed idea can improve the results as compared to the default Apriori algorithm.
- 2:40 A Text Mining Approach for Mapping OCL Navigation Expressions
-
One problem of exploiting information captured by an OO system model such as class models is mapping user queries on the model. This is particularly due to the informal character of the user queries. This paper presents an approach based on text mining for identifying and interpreting user queries. This would enable a matching between the interpreted user query and the corresponding entities specified by the system model. Now navigation expressions in OCL can be derived from the system model that have a formal character and are able to answer the user questions against the class model.
- 3:00 Restructuring Concept Based Censor Production Rules
-
In this paper we present a new structure for the Concept Based Censor Production Rule (CBCPR). CBCPR is an extension of a rule called censor production rule. CBCPR and censor production rules can be used in real time systems. Censor production rule lacks the existence of what so called concept. This lacking might make the matching and firing of the rules take quite some more time. To avoid this problem, CBCPR has been introduced to include the idea of concept in its structure. Despite that CBCPR solved the problem, its structure needs to be restructured so that it can put more than one concept related to similar conditions in one structure. This restructuring makes the writing of CBCPR easier and it also expedites the execution process. Utilizing the given time in a better way means giving more opportunities to check more censor conditions and produce a more certain result. The proposed restructuring will also make the computation easier and faster for the certainty values for various concepts related to one rule structure.
S18: Informatics-3
Room: S45-11-Zain E-Learning Center
Chairs: Ali H Zolait, AZ, Faisal Hammad
- 2:00 Analysis of Customers Comments on Social Media Websites of Cosmetic Brands
-
In this article we discuss several techniques of sentiment analysis and compare them with social media contents from business competitors. User generated contents in a very big number is freely available on different social media sites now a day. Mostly companies increase their competitive advantages keep an eye on their competing companies and closely analyze the data that are generated by their customers on their social media sites. In this article I integrate the several techniques using a framework to analyze and make a comparison between social media content and business competitors. These techniques include the competitive analysis, text mining and quantitative analysis. Specifically, this article is going to analyze the three big brands of cosmetics (MAC, soft touch, Nivea, Revlon and L'Oreal) and will compare the competitive analysis among them on social media sites. When I analyzing these five big brands I found some similarities among their social media usage. This article discusses the suggestions of my discussions and provides the strong recommendations for helping companies to build their strong competitive analysis strategies on social media. We discuss the implications of our findings and provide recommendations to help companies develop their social media competitive analysis strategies.
- 2:20 Usability of the Academic Transcript
-
In this paper we study the usability of the text-based online students' transcripts. The usability test depends on the effectiveness, efficiency, control, and learnability usability characteristics of the academic transcript. The user experience and the aesthetic Nielsen design heuristic were studied as well. The usability test criteria designed to fit each usability feature. Those include the ability to plan for upcoming semesters, finding the major GPA, customizing the displayed information, easiness of use, and attractiveness.
- 2:40 A Tutoring Web Service for E-Learning System
-
The service-oriented architecture (SOA), continues to gain ground as an approach that enables agility, flexibility and cost savings. It has been adopted by different IT communities. The field of e-learning was no exception to the rule, as a multitude of work for the adoption of SOA was proposed. In a preliminary study we conducted, we focused our efforts on proposing a global service-oriented architecture for e-learning systems that promotes the reuse of its various functionalities as remote modules (services). In this article, we focus on the tutoring aspect of elearning systems. The idea is to offer and make available a web service that is not an integral part of the e-learning system, but remains independent as a web service, this independence allowing it to extend any e-learning platform with the tutoring function. This service has been designed to meet SOA requirements in terms of autonomy, reuse, interoperability and remote access. Offering the tutoring aspect as an independent service allows the different existing platforms without this functionality to benefit from it without making major changes to the code, especially since SOA standards allow easy communication between heterogeneous platforms developed in different languages and systems. In order to validate our approach, we developed and tested the tutoring web service. It is a tool that allows the tutor to closely monitor the progress of each learner in his studies, in his collaboration and in his different interactions with other learners and with the tutor. Many other statistics through this web service, which allows the tutor to have an idea about the seriousness of the learner and to predict the risk of dropping out of courses.
- 3:00 Utilization of Social Media for Rural Women Entreprenuer Project: A Case Study of Setiu Wetland
-
Entrepreneurship empowerment towards rural women remains crucial to attain better living standard among rural communities. To help improve the living livelihood through e-business, the community needs to be equipped with entrepreneurship skill explicitly and Information and Communication (ICT) skill implicitly. This paper presents a study case of rural women entrepreneur in Setiu Wetland, Terengganu, Malaysia under NRGS research grant from the period of 2015 to 2018 by a group of UMT researchers and academicians. The study utilizes facebook social media as the electronic business (e-business) and networking platform in marketing and establishing the local products of Setiu Wetland women entrepreneurs. Through the six (6) months period of tracking the respondent's income rate, the outcome of this NRGS project depicts 17% of women entrepreneurs achieve linear proportion of increasing income rate, 33% of decreasing income rate and the rest 50% of income shows a fluctuating rate.
Monday, November 19 3:20 - 3:45
CS: Closing Session
Room: S45-101-Zain E-Learning Center
Chair: Faisal Hammad