M. BARKAT Abdelbasset

MCA

Directory of teachers

Department

Informatics Department

Research Interests

Specialized in Informatics Department. Focused on academic and scientific development.

Contact Info

University of M'Sila, Algeria

Email : abdelbasset.barkat@univ-msila.dz

On the Web:

  • Google Scholar N/A
  • ResearchGate
    ResearchGate N/A
  • ORCID N/A
  • SC
    Scopus N/A

Recent Publications

2025-05-22

Urban Traffic Prediction Using Hybrid XGBoost–LSTM Model

This study introduces a novel hybrid predictive model that integrates eXtreme Gradient Boosting (XGBoost) with Long Short-Term Memory (LSTM) networks, specifically designed for real-time urban traffic congestion prediction. The proposed model innovatively incorporates external data, such as weather conditions, traffic incidents, and road classifications, and effectively addresses the common issue of class imbalance in the traffic dataset and captures dynamic spatiotemporal traffic relationships. This comprehensive approach enables the capture of complex, dynamic spatiotemporal traffic relationships more accurately. XGBoost performs robust feature selection and preliminary classification, generating probabilistic traffic jam level estimates. These outputs are subsequently enhanced through bidirectional LSTM layers that leverage temporal dependencies within traffic data, thus significantly improving predictive accuracy. The hybrid XGBoost–LSTM model was evaluated using approximately three million real-time traffic records from central London, providing a substantial and realistic testing environment. The results demonstrated its superior performance, achieving an accuracy of 93%, with precision values between 86% and 96%, and recall between 84% and 97% across varying congestion scenarios, from free flow to heavy congestion. Notably, the inclusion of probabilistic feature augmentation successfully mitigated the impact of class imbalance, further enhancing reliability. Comparative analyses against traditional and standalone methods highlighted the proposed hybrid model’s substantial improvement in accurately differentiating traffic jam levels, making it a valuable tool for intelligent transportation systems (ITS). This research contributes significantly to urban traffic management strategies, supporting smoother traffic flow and congestion reduction.
Citation

M. BARKAT Abdelbasset, Derya Yiltas-Kaplan, , (2025-05-22), "Urban Traffic Prediction Using Hybrid XGBoost–LSTM Model", [national] International Journal of Computing and Digital Systems , scopus

Read Full Paper
2024-12-11

Real-Time Data Integration for Effective Congestion Forecasting

Congestion in smart cities is a challenging issue due to its impact on people’s lives. Therefore, many researchers
are developing solutions to this problem by taking advantage of new technologies, especially sensors and wireless
network infrastructures. These technologies enable the collection of real-time information about many features
directly or indirectly linked to traffic congestion.
In this paper, we propose a model for traffic congestion prediction based on real-time data, considering a set of
relevant features such as traffic flow, incidents, holidays, and weather data. Our prediction classifies data into four
categories: free flow, mild congestion, moderate congestion, and heavy congestion. After analyzing the results, the
algorithm with the highest performance is XGBoost, achieving a prediction accuracy of 90%, followed by random
forest and decision tree, both with a prediction accuracy of 89%
Citation

M. BARKAT Abdelbasset, (2024-12-11), "Real-Time Data Integration for Effective Congestion Forecasting", [international] The Sixth International Symposium on Informatics and Its Applications (ISIA) , M'sila

Read Full Paper
2024-06-26

Enhanced Method in Artificial Intelligence and Machine Learning for Enhanced Computer Vision Application

For computers to be able to see, researchers are studying computer vision. Even the most general computer vision problems include drawing conclusions about the environment from images. It draws from a variety of disciplines and might be considered a subfield of AI and ML, which employ both generalizable and domain-specific learning strategies. The use of techniques from other fields, such as computer science and engineering, can make interdisciplinary research appear disorganized. A sophisticated ensemble of generic machine learning algorithms can be needed to handle a different visual problem than a hand-crafted statistical technique. Modern science has revolutionized computer vision. Exciting and often chaotic, frontier areas often have few trustworthy authorities. Some theories work in theory but not in practice, while many excellent ideas are theoretically unfounded. A lot of the developed world is spread out and can look like it's out of reach. These days, deep learning, machine learning, and computer vision all work well. The cornerstones of any school are its teaching and learning programmes. Students' actions and presence in class are closely observed alongside their academic progress. Classroom monitoring, emotion recognition, appraisal, and real-time attendance tracking were some of the computer vision applications studied in this research. A wide range of viewpoints have explored computer vision. Digital picture processing, pattern identification, machine learning, and computer graphics are all a part of it, in addition to raw data recording. Because of its extensive application, many researchers include it into a wide range of disciplines.
Citation

M. BARKAT Abdelbasset, (2024-06-26), "Enhanced Method in Artificial Intelligence and Machine Learning for Enhanced Computer Vision Application", [national] INTELLIGENT SYSTEMS AND APPLICATIONS IN ENGINEERING , Auricle Global Society of Education and Research

Read Full Paper
2023-05-23

Extracting sequential frequent itemsets from probabilistic sequences database

omputers now handle large amounts of data, leading to the emergence of data mining as a science to extract useful information from this data. Frequent itemset mining is a popular technique used to discover relationships between items in big data. However, in real-life scenarios, data may be incomplete or uncertain, posing a challenge for frequent itemset mining. This paper proposes a novel approach for mining sequential frequent itemsets from uncertain sequence databases. The approach comprises two main parts: extracting a set of probabilistic frequent itemsets, and filtering this set using the sequential information in the data. At the end of the paper, we provide a comparison with an existing method to further demonstrate the value of our approach.
Citation

M. BARKAT Abdelbasset, (2023-05-23), "Extracting sequential frequent itemsets from probabilistic sequences database", [national] International Journal of Information Technology , springer

Read Full Paper
2021

Framework for web service composition based on QoS in the multi cloud environment

Recently, the term cloud computing is widely used in the searching community, which shows the importance given by the scientists to this research area; cloud computing is a new computing model provides shared resources and data based on service delivery model, where everything from infrastructures, platforms, and software are given to the user like a set of services. The users of cloud platforms deal with these services to satisfy their requests, however these requests become more complex, and they need more than one service to accomplish one request; the process of gathering a set of services to satisfy a user request is called the service composition. In addition to the fact that one request needs a set of services to be executed, and die to the quick development of cloud technology, there are many of similar services which offer the same functionality, for each service in this set, which make the composition process needs a mechanism to choose between these infinity choices to give the user an optimal satisfaction. In this paper, We propose a framework for service composition in the multi cloud environment where we compose these services based on two factors: the first is a set of QoS (quality of service) criteria for each service, and the second is the number of cloud bases involved in the composition process, to build a composed service that satisfy the user request.
Citation

M. BARKAT Abdelbasset, Kazar Okba, Imane SEDDIKI, , (2021), "Framework for web service composition based on QoS in the multi cloud environment", [national] International Journal of Information Technology , springer

Read Full Paper
2019

Security model based mobile agent for mobile ad hoc networks

In this paper, we propose a security model based mobile agent for MANETs, the objective here is to improve the security level of the network communication without affecting its performance. The model includes a hierarchical clustering and mobile agents. We apply the concept of dominating set based clustering for partitioning the network into clusters. The cluster head election is based on both the trust and resources ability of the node. We define four agent types. The node agent manages the use of node resources. The ambassador agent is created by the control agent to monitor all the actions of the node agent. The control agent is created in the most trusted with best resources node to control the communication into the cluster and participates with its counterparts in the security network completely, and the transporter agent carries the encrypted information in the network.
Citation

M. BARKAT Abdelbasset, (2019), "Security model based mobile agent for mobile ad hoc networks", [national] International Journal of Communication Networks and Distributed Systems , Inderscience Publishers (IEL)

Read Full Paper

Smart and fuzzy approach based on CSP for cloud resources allocation

In this study, we proposed a resource allocation approach that aims at fulfilling two main objectives. First, it equilibrates between the different cloud infrastructure particularities including load balancing, so it enhances the performance of infrastructure. Second, our approach provides a solution for the customer needs through shortening the execution time and reducing payments of the requested resources that have a dynamic nature. This paper suggests a new hybrid resources allocation approach based on three methods: multiagent system (MAS), distributed constraints satisfaction problems (DCSP), and the fuzzy logic (FL). The MAS represents the physical infrastructure. It provides an efficient management of the resources in the distribution and the heterogeneity of this infrastructure. DCSP, on the other hand, works side by side with MAS to maintain resources allocation policies in datacenters, while the FL is used to facilitate the representation of the dynamic resource values into linguistic terms (low, medium, high  …) and helps the system to determine the best solution according to the criteria of customer requests. The experimental results show the efficiency of our approach in terms of load balancing, energy consumption cost, execution time, and the rate payment gain of customers.
Citation

M. BARKAT Abdelbasset, (2019), "Smart and fuzzy approach based on CSP for cloud resources allocation", [national] International Journal of Computers and Applications , Taylor & Francis

Read Full Paper
2018

Service composition in the multi cloud environment

Purpose
User requests over the cloud are not achievable with one single service, multiple services need to be executed to fulfill what a user asks for. Typically, such services are composed and presented as one global service. Moreover, the same operation can be achieved by multiple services available at different clouds, which can result in different possibilities in composing them. This paper aims to decrease the number of clouds involved in the composition process, so that user requests are satisfied with minimal cost (communication costs, execution time and financial charges).

Design/methodology/approach
This paper investigates the use of an intelligent water drops (IWDs) optimization-based algorithm, and an integer linear programming model to optimize the number of cloud bases involved in the composition process. A comparison of the solutions found by these two techniques is presented in the paper.

Findings
The obtained results show that the number of cloud bases can be decreased without affecting user satisfaction.

Originality/value
The paper is a first attempt to use the IWDs algorithm for service composition, tested with big-size data
Citation

M. BARKAT Abdelbasset, (2018), "Service composition in the multi cloud environment", [national] International Journal of Web Information Systems , Emerald Publishing Limited

Read Full Paper
2013

Agent-based approach for building ontology from text

An ontology is an explicit specification of a conceptualization, the term is often linked with the Semantic Web, ontologies are used like representations of knowledge, to annotate web resources and also to communication between systems. This make them very important, but unfortunately their construction is expensive, and because they are representations of knowledge, we thought of using the enormous amount of information available under textual format to automate the process of ontology building, and since we deal with texts, the NLP (Natural Language Processing) is considered as the base for the ontology construction from text. In this paper our goal is to propose an agent-based approach to build ontology from text, and implement a multi-agent system guided by this approach, which start from a set of textual resources to give us an ontology in OWL (Ontology Web Language), using the Formal Concept Analysis FCA and Relational Concept Analysis RCA to move from the syntactic level to the semantic level.
Citation

M. BARKAT Abdelbasset, (2013), "Agent-based approach for building ontology from text", [international] International Conference on Computer Applications Technology (ICCAT) , Sousse

Read Full Paper
← Back to Researchers List