Syntactic Annotation in the I3rab Dependency

Treebank

Dana Halabi¹, Arafat Awajan^1,3, and Ebaa Fayyoumi²

¹Department of Computer Science, Princess Sumaya University for Technology, Jordan

²Department of Computer Science, Hashemite University, Jordan

³Information Technology College, Computer Science Department, Mutah University, Jordan

Abstract: Arabic dependency parsers have a poor performance compared to parsers of other languages. Recently the impact of annotation at lexical level of dependency treebank on the overall performance of the dependency parses has been extensively investigated. This paper focuses on the impact of coarse-grained and fine-grained dependency relations on the performance of Arabic dependency parsers. Moreover, this paper introduces the annotation rules for I3rab dependency treebank. Experimentally, the obtained results showed that having an appropriate set of dependency relations improves the performance of an Arabic dependency parser up to 27.55%.

Keywords: Arabic language, dependency parsing, natural language processing, dependency structure, dependency relation, annotation rules.

Received February 20, 2021; accepted March 7, 2021

https://doi.org/10.34028/iajit/18/3A/1

Full text   

Multi-Agents Collaboration in Open System

Zina Houhamdi¹ and Belkacem Athamena²

¹Software Engineering Department, College of Engineering, Al Ain University, UAE

²Business Administration Department, College of Business, Al Ain University, UAE

Abstract: Share constrained resources, accomplish complex tasks and achieve shared or individual goals are examples requiring collaboration between agents in multi-agent systems. The collaboration necessitates an effective team composed of a set of agents that do not have conflicting goals and express their willingness to cooperate. In such a team, the complex task is split into simple tasks, and each agent performs its assigned task to contribute to the fulfilment of the complex task. Nevertheless, team formation is challenging, especially in an open system that consists of self-interested agents performing tasks to achieve several simultaneous goals, usually clashing, by sharing constrained resources. The clashing goals obstruct the collaboration's success since the self-interested agent prefers its individual goals to the team’s shared goal. In open systems, the collaboration team construction process is impacted by the Multi-Agent System (MAS) model, the collaboration’s target, and dependencies between agents’ goals. This study investigates how to allow agents to build collaborative teams to realize a set of goals concurrently in open systems with constrained resources. This paper proposes a fully distributed approach to model the Collaborative Team Construction Model (CTCM). CTCM modifies the social reasoning model to allow agents to achieve their individual and shared goals concurrently by sharing resources in an open MAS by constructing collaborative teams. Each agent shares partial information (to preserve privacy) and models its goal relationships. The proposed team construction approach supports a distributed decision-making process. In CTCM, the agent adapts its self-interest level and adjusts its willingness to form an effective collaborative team.

Keywords: Multi-agents system; open system; collaboration, dependency relationships, decentralized decision making.

Received February 20, 2021; accepted March 7, 2021

https://doi.org/10.34028/iajit/18/3A/2 

Full text  

Multi-Spectral Hybrid Invariant Moments Fusion Technique for Face Identification

Shaymaa Hamandi, Abdul Monem Rahma, and Rehab Hassan

Computer Science Department, University of Technology, Iraq

Abstract: For reliable face identification, the fusion process of multi-spectral vision features produces robust classification systems, this paper exploits the power of thermal facial image invariant moments features fused with the visible facial image invariant moments features to propose a new multi-spectral hybrid invariant moment fusion system for face identification. And employs Feed-forward neural network to train the moments' features and make decisions. The evaluation system uses databases of visible thermal pairs face images CARL and UTK-IRIS databases and gives an accuracy reaches 99%.

Keywords: Feed-forward neural network, affine moments, face recognition, invariant moments, shape descriptor, zernike moments.

Received February 21, 2021; accepted March 7, 2021

Usability Study of Enhanced Salat Learning Approach using Motion Recognition System

Nor Azrini Jaafar, Nor Azman Ismail, and Yusman Azimi Yusoff

Faculty of Engineering, Universiti Teknologi Malaysia, Malaysia

Abstract: Salat learning is one of the most important processes for every Muslim. The current learning approach requires the teacher or expert to be present in the learning session, which is time and resource-consuming. Previous researchers use both wearable and non-wearable sensors to recognize salat movement. They focus on specific salat movement rather than complete salat movement cycle. This paper present a motion recognition system to enhance salat learning experience. The system helps users recognize the complete salat movements cycle based on salat law by using a multisensor setup for better tracking capability. Three evaluations are conducted to validate the system's performance and its contribution. The first evaluation is to measure success score in recognition accuracy and identify user error. The second evaluation is conducted to compare the proposed system with the traditional-based methodology, and the last evaluation focuses on the user experience and acceptance of the proposed system. The result from performance evaluation shows the system has high accuracy in recognizing salat movement. There is a significant difference in the error rate and success score when comparing the learning methodology. However, users provide positive feedback based on the survey conducted after using the proposed system.

Keywords: Salat Learning, motion recognition, salat law, learning methodology.

Received February 20, 2021; accepted March 7, 2021

https://doi.org/10.34028/iajit/18/3A/4

Full text     

Abstract: With latest development in technology, the usage of smartphones to fulfill day-to-day requirements has been increased. The Android-based smartphones occupy the largest market share among other mobile operating systems. The hackers are continuously keeping an eye on Android-based smartphones by creating malicious apps housed with ransomware functionality for monetary purposes. Hackers lock the screen and/or encrypt the documents of the victim’s Android based smartphones after performing ransomware attacks. Thus, in this paper, a framework has been proposed in which we (1) utilize novel features of Android ransomware, (2) reduce the dimensionality of the features, (3) employ an ensemble learning model to detect Android ransomware, and (4) perform a comparative analysis to calculate the computational time required by machine learning models to detect Android ransomware. Our proposed framework can efficiently detect both locker and crypto ransomware. The experimental results reveal that the proposed framework detects Android ransomware by achieving an accuracy of 99.67% with Random Forest ensemble model. After reducing the dimensionality of the features with principal component analysis technique; the Logistic Regression model took least time to execute on the Graphics Processing Unit (GPU) and Central Processing Unit (CPU) in 41 milliseconds and 50 milliseconds respectively.

Keywords: Smartphone security, android, ensemble learning, ransomware, and dimensionality reduction.

Jennifer Batamuliza¹ and Damien Hanyurwimfura²

¹College of Business and Economics, University of Rwanda, Rwanda

²College of Science and Technology, University of Rwanda, Rwanda

Abstract: Smart Grid (SG) is a modern digital metering system that was introduced by researchers to take over the traditional electricity infrastructure that existed before by gathering and putting in use the data generated by smart meters and ensure efficiency and reliability in the two directional flow of electricity and data for both the service providers and smart meters. Leakage of customers’ identity causes inconvenience to the customer because he is exposed to theft in his household. Secure anonymous key distribution scheme for SG has been proposed as solution to secure data transfer between service provider and customer. Existing secure anonymous key distribution scheme for SG brings challenge such as being inefficient, having traceability issues and do not stop replay attack hence vulnerable to DoS attacks. In this paper a Secure efficient anonymous identity-based with bilateral protocol is proposed to address the weakness in existing anonymous key distribution schemes. , With this protocol, both smart meter and service provider in (SG) identify each other anonymously in efficient way achieving un-traceability and restisting Replay and DoS attack.

Keywords: Identity-based, anonymous, bilateral protocol, smart grid, smart meters.

A Personalized Recommendation for Web API Discovery in Social Web of Things

Marwa Meissa¹, Saber Benharzallah², Laid Kahloul¹, and Okba Kazar¹

¹LINFI Laboratory, Biskra University, Algeria

²Department of Computer Science, LAMIE Laboratory Batna 2 University, Algeria

Abstract: With the explosive growth of Web of Things (WoT) and social web, it is becoming hard for device owners and users to find suitable web Application Programming Interface (API) that meet their needs among a large amount of web APIs. Social-aware and collaborative filtering-based recommender systems are widely applied to recommend personalized web APIs to users and to face the problem of information overload. However, most of the current solutions suffer from the dilemma of accuracy- diversity where the prediction accuracy gains are typically accompanied by losses in the diversity of the recommended APIs due to the influence of popularity factor on the final score of APIs (e.g., high rated or high-invoked APIs). To address this problem, the purpose of this paper is developing an improved recommendation model called (Personalized Web API Recommendation) PWR, which enables to discover APIs and provide personalized suggestions for users without sacrificing the recommendation accuracy. To validate the performance of our model, seven variant algorithms of different approaches (popularity-based, user-based and item-based) are compared using MovieLens 20M dataset. The experiments show that our model improves the recommendation accuracy by 12% increase with the highest score among compared methods. Additionally it outperforms the compared models in diversity over all lengths of recommendation lists. It is envisaged that the proposed model is useful to accurately recommend personalized web API for users.

Keywords: Web of Things, recommender system, web API, collaborative filtering, rating prediction, social networks, IoT.

Received February 20, 2021; accepted March 7, 2021

 https://doi.org/10.34028/iajit/18/3A/7

Full text      

Generating Sense Inventories for

Ambiguous Arabic Words

Marwah Alian¹ and Arafat Awajan^1,2

¹King Hussein School of Computing Sciences, Princess Sumaya University for Technology, Jordan

²Information Technology College, Computer Science Department, Mutah University, Jordan

Abstract: The process of selecting the appropriate meaning of an ambigous word according to its context is known as word sense disambiguation. In this research, we generate a number of Arabic sense inventories based on an unsupervised approach and different pre-trained embeddings, such as Aravec, Fasttext, and Arabic-News embeddings. The resulted inventories from the pre-trained embeddings are evaluated to investigate their efficiency in Arabic word sense disambiguation and sentence similarity. The sense inventories are generated using an unsupervised approach that is based on a graph-based word sense inductionalgorithm. Results show that the Aravec-Twitter inventory achieves the best accuracy of 0.47 for 50 neighbors and a close accuracy to the Fasttext inventory for 200 neighbors while it provides similar accuracy to the Arabic-News inventory for 100neighbors. The experiment of replacing ambiguous words with their sense vectors is tested for sentence similarity using all sense inventories and the results show that using Aravec-Twitter sense inventoryprovides a better correlation value.

Keywords: Word sense induction, word sense disambiguation, arabic text, sense inventory.

Received February 25, 2021; accepted March 7, 2021

Keywords: Capacitated vehicle routing problem, supervised fuzzy c-means, fuzzy clustering.

Received February 20, 2021; accepted March 7, 2021

https://doi.org/10.34028/iajit/18/3A/9

Full text    

A Multi-Population Genetic Algorithm for Adaptive QoS-Aware Service Composition in Fog-IoT Healthcare Environment

Idir Aoudia¹, Saber Benharzallah², Laid Kahloul¹, and Okba Kazar¹

¹LINFI Laboratory, Biskra University, Algeria

² LAMIE Laboratory, Computer Sciences department, Batna 2 University, Algeria

Abstract: The growth of Internet of Thing (IoT) implies the availability of a very large number of services which may be similar or the same, managing the Quality of Service (QoS) helps to differentiate one service from another. The service composition provides the ability to perform complex activities by combining the functionality of several services within a single process. Very few works have presented an adaptive service composition solution managing QoS attributes, moreover in the field of healthcare, which is one of the most difficult and delicate as it concerns the precious human life.In this paper, we will present an adaptive QoS-Aware Service Composition Approach (P-MPGA) based on multi-population genetic algorithm in Fog-IoT healthcare environment. To enhance Cloud-IoT architecture, we introduce a Fog-IoT 5-layared architecture. Secondly, we implement a QoS-Aware Multi-Population Genetic Algorithm (P-MPGA), we considered 12 QoS dimensions, i.e., Availability (A), Cost (C), Documentation (D), Location (L), Memory Resources (M), Precision (P), Reliability (R), Response time (Rt), Reputation (Rp), Security (S), Service Classification (Sc), Success rate (Sr), Throughput (T). Our P-MPGA algorithm implements a smart selection method which allows us to select the right service. Also, P-MPGA implements a monitoring system that monitors services to manage dynamic change of IoT environments. Experimental results show the excellent results of P-MPGA in terms of execution time, average fitness values and execution time / best fitness value ratio despite the increase in population. P-MPGA can quickly achieve a composite service satisfying user’s QoS needs, which makes it suitable for a large scale IoT environment.

Keywords: IoT, service composition, adaptability, context; QoS, Fog-IoT computing, Healthcare.

Received February 20, 2021; accepted March 7, 2021

https://doi.org/10.34028/iajit/18/3A/10

Full text   

LoRaWAN Energy Optimization with Security

Consideration

Ala Khalifeh¹, Khaled Aldahdouh¹ and Sahel Alouneh^1,2

¹School of Electrical Engineering and Information Technology, German Jordanian University, Jordan

²College of Engineering, Al Ain University, UAE

Abstract: Long Range Wide Area Network (LoRaWAN) is an emerging wireless technology that is expected to be widely deployed and implemented in several applications, especially with the promising widespread use of the Internet of Things (IoT) and its potential applications within the Fifth Generation (5G) communication technology. LoRaWAN consists of a number of nodes that monitors and senses the environment to collect specific data, and then sends the collected data to a remote monitoring device for further processing and decision-making. Energy consumption and security assurance are two vital factors needed to be optimized to ensure an efficient and reliable network operation and performance. To achieve that, each of LoRaWAN nodes can be configured by five transmission parameters, which are the spreading factor, carrier frequency, bandwidth, coding rate and transmission power. Choosing the best values of these parameters leads to enhancing the network deployment. In this paper, we shed the light to the security aspect in LoRaWAN network. Then, we introduced an algorithm that depends on the reinforcement learning approach to enable each node in the network to choose the best values of spreading factor and transmission power such that it leads to a lower energy consumption and higher packets’ delivery rate. The results of the simulation experiments of our proposed technique showed a valuable increase in the packet reception rate at the gateway and a significant decrease in the total consumed energy at the end nodes compared with the most related work in literature.

Keywords: LoRaWAN, transmission parameters, reinforcement learning, power consumption, Security, Confidentiality, Authentication.

Received February 28, 2021; accepted March 7, 2021

 https://doi.org/10.34028/iajit/18/3A/11

Full text  

Computer Vision in Contactless Biometric

Systems

Farukh Hashmi¹, Kiran Ashish², Satyarth Katiyar³, and Avinash Keskar⁴

¹Department of Electronics and Communication Engineering, National Institute of Technology, India

²Computer Vision Engineer, Viume, India

³Department of Electronics and Communication Engineering, Harcourt Butler Technical University, India

⁴Department of Electronics and Communication Engineering, Visvesvaraya National Institute of Technology, India

Abstract: Contactless biometric systems have increased ever since the corona pandemic outbreak. The two main contactless biometric systems are facial recognition and gait patterns recognition. The authors in the previous work [11] have built hybrid architecture AccessNet. It involves combination of three systems: facial recognition, facial anti-spoofing, and gait recognition. This work involves deploying the hybrid architecture and deploying two individual systems such as facial recognition with facial anti-spoofing and gait recognition individually and comparing the individual results in real-time with the AccessNet hybrid architecture results. This work even involves in identifying the main crucial features from each system that are responsible for predicting a subject. It includes extracting few crucial parameters from gait recognition architecture, facial recognition and facial anti-spoof architectures by visualizing the hidden layers. Each individual method is trained and tested in real-time, which is deployed on both edge device NvidiaJetsonNano, and high-end GPU. A conclusion is also adapted in terms of commercial and research usage for each single method after analysing the real-time test results.

Keywords: AccessNet, gait patterns, facial recognition, contactless biometric systems, crucial features, NvidiaJetsonNano.

Received February 21, 2021; accepted March 7, 2021

https://doi.org/10.34028/iajit/18/3A/12

Full text