Kamel Boukhalfa
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Research Interests
Kamel Boukhalfa has a total of 33 co-authors. Here are the top co-authors by number of shared publications:
- Rahma Djiroun
Department of Computer Science, LSI/USTHB, Algiers, Algeria
3 shared publications - Mohammed Brahimi
Intelligent Systems Enginnering, National School of Artificial Intelligence (ENSIA), Algiers, Algeria
2 shared publications Pascal Richard
University of Poitiers, France
2 shared publications- Sami Belkacem
LSI Laboratory, USTHB
2 shared publications - Sandro Bimonte
TSCF, INRAE - University of Clermont Auvergne
2 shared publications Abdelouahab Moussaoui
Department of Computer Science, Setif 1 University, Setif, Algeria
1 shared publication
Latest Publications
- 17 May 2024
- Published by Springer Nature in Cluster Computing
- Vol. 27 (8) , 1-29
Natural language why-question answering system in business intelligence context
- 17 May 2024
- Published by Springer Nature in Cluster Computing
- Vol. 27 (8) , 1-29
Natural language why-question answering system in business intelligence context
- 28 September 2022
- Published by Springer Nature
- p. 15-26
Causality Analysis Method and Model Related to Why-Question Answering in Business Intelligence Context
- 28 September 2022
- Published by Springer Nature
- p. 15-26
Causality Analysis Method and Model Related to Why-Question Answering in Business Intelligence Context
- 1 October 2021
- Published by IGI Global Scientific Publishing in International Journal of Information Technology and Web Engineering
- Vol. 16 (4) , 52-73
An Improved Sentiment Analysis Approach to Detect Radical Content on Twitter
Social networks are used by terrorist groups and people who support them to propagate their ideas, ideologies, or doctrines and share their views on terrorism. To analyze tweets related to terrorism, several studies have been proposed in the literature. Some works rely on data mining algorithms; others use lexicon-based or machine learning sentiment analysis. Some recent works adopt other methods that combine multi-techniques. This paper proposes an improved approach for sentiment analysis of radical content related to terrorist activity on Twitter. Unlike other solutions, the proposed approach focuses on using a dictionary of weighted terms, the Word2vec method, and trigrams, with a classification based on fuzzy logic. The authors have conducted experiments with 600 manually annotated tweets and 200,000 automatically collected tweets in English and Arabic to evaluate this approach. The experimental results revealed that the new technique provides between 75% to 78% of precision for radicality detection and 61% to 64% to detect radicality degrees.
- 1 October 2021
- Published by IGI Global Scientific Publishing in International Journal of Information Technology and Web Engineering
- Vol. 16 (4) , 52-73
An Improved Sentiment Analysis Approach to Detect Radical Content on Twitter
Social networks are used by terrorist groups and people who support them to propagate their ideas, ideologies, or doctrines and share their views on terrorism. To analyze tweets related to terrorism, several studies have been proposed in the literature. Some works rely on data mining algorithms; others use lexicon-based or machine learning sentiment analysis. Some recent works adopt other methods that combine multi-techniques. This paper proposes an improved approach for sentiment analysis of radical content related to terrorist activity on Twitter. Unlike other solutions, the proposed approach focuses on using a dictionary of weighted terms, the Word2vec method, and trigrams, with a classification based on fuzzy logic. The authors have conducted experiments with 600 manually annotated tweets and 200,000 automatically collected tweets in English and Arabic to evaluate this approach. The experimental results revealed that the new technique provides between 75% to 78% of precision for radicality detection and 61% to 64% to detect radicality degrees.
- 1 January 2021
- Published by IGI Global Scientific Publishing in International Journal of Data Warehousing and Mining
- Vol. 17 (1) , 15-38
Ranking News Feed Updates on Social Media: A Review and Expertise-Aware Approach
Social media are used by hundreds of millions of users worldwide. On these platforms, any user can post and share updates with individuals from his social network. Due to the large amount of data, users are overwhelmed by updates displayed chronologically in their newsfeed. Moreover, most of them are irrelevant. Ranking newsfeed updates in order of relevance is proposed to help users quickly catch up with the relevant updates. In this work, the authors first study approaches proposed in this area according to four main criteria: features that may influence relevance, relevance prediction models, training and evaluation methods, and evaluation platforms. Then the authors propose an approach that leverages another type of feature which is the expertise of the update's author for the corresponding topics. Experimental results on Twitter highlight that judging expertise, which has not been considered in the academic and the industrial communities, is crucial for maximizing the relevance of updates in newsfeeds.
- 1 January 2021
- Published by IGI Global Scientific Publishing in International Journal of Data Warehousing and Mining
- Vol. 17 (1) , 15-38
Ranking News Feed Updates on Social Media: A Review and Expertise-Aware Approach
Social media are used by hundreds of millions of users worldwide. On these platforms, any user can post and share updates with individuals from his social network. Due to the large amount of data, users are overwhelmed by updates displayed chronologically in their newsfeed. Moreover, most of them are irrelevant. Ranking newsfeed updates in order of relevance is proposed to help users quickly catch up with the relevant updates. In this work, the authors first study approaches proposed in this area according to four main criteria: features that may influence relevance, relevance prediction models, training and evaluation methods, and evaluation platforms. Then the authors propose an approach that leverages another type of feature which is the expertise of the update's author for the corresponding topics. Experimental results on Twitter highlight that judging expertise, which has not been considered in the academic and the industrial communities, is crucial for maximizing the relevance of updates in newsfeeds.
- 1 November 2019
- Published by Springer Nature in Cluster Computing
- Vol. 23 (3) , 2375-2388
Expertise-aware news feed updates recommendation: a random forest approach
- 1 November 2019
- Published by Springer Nature in Cluster Computing
- Vol. 23 (3) , 2375-2388
Expertise-aware news feed updates recommendation: a random forest approach
- 1 July 2019
- Published by IGI Global Scientific Publishing in International Journal of E-Business Research
- Vol. 15 (3) , 16-34
Towards the Meta-Modeling of Complex Inter-Organisationnel Collaborative Processes
Enterprises are progressively embedded in business to business atmospheres, in order to achieve their common business objectives. Such collaborations lead to Inter-Organizational Business Processes. Therefore, IOBP modeling involves new challenges, mainly the ability to deal with autonomy, privacy, heterogeneity. As a contribution in this area, a IOBP meta-model was designed. This model takes into account the maximum concepts related to the collaboration. Where, the process is complex, and its model in a global way affects its vision and complicates its implementation and hence the idea of its analysis into sub-IOBP to reduce the complexity of the global collaborative process, to streamline information exchange and to facilitate the understanding of the process by partners. A set of Atlas Transformation Language (ATL) transformation rules has been defined to convert Unified Modeling Language (UML) models to Business Process Model and Notation. Finally, the application of our approach has been demonstrated through a framework which can solve the problem of generic IOBP.
- 1 July 2019
- Published by IGI Global Scientific Publishing in International Journal of E-Business Research
- Vol. 15 (3) , 16-34
Towards the Meta-Modeling of Complex Inter-Organisationnel Collaborative Processes
Enterprises are progressively embedded in business to business atmospheres, in order to achieve their common business objectives. Such collaborations lead to Inter-Organizational Business Processes. Therefore, IOBP modeling involves new challenges, mainly the ability to deal with autonomy, privacy, heterogeneity. As a contribution in this area, a IOBP meta-model was designed. This model takes into account the maximum concepts related to the collaboration. Where, the process is complex, and its model in a global way affects its vision and complicates its implementation and hence the idea of its analysis into sub-IOBP to reduce the complexity of the global collaborative process, to streamline information exchange and to facilitate the understanding of the process by partners. A set of Atlas Transformation Language (ATL) transformation rules has been defined to convert Unified Modeling Language (UML) models to Business Process Model and Notation. Finally, the application of our approach has been demonstrated through a framework which can solve the problem of generic IOBP.
- Open Access
Deep interpretable architecture for plant diseases classification
Recently, many works have been inspired by the success of deep learning in computer vision for plant diseases classification. Unfortunately, these end-to-end deep classifiers lack transparency which can limit their adoption in practice. In this paper, we propose a new trainable visualization method for plant diseases classification based on a Convolutional Neural Network (CNN) architecture composed of two deep classifiers. The first one is named Teacher and the second one Student. This architecture leverages the multitask learning to train the Teacher and the Student jointly. Then, the communicated representation between the Teacher and the Student is used as a proxy to visualize the most important image regions for classification. This new architecture produces sharper visualization than the existing methods in plant diseases context. All experiments are achieved on PlantVillage dataset that contains 54306 plant images.
Open AccessDeep interpretable architecture for plant diseases classification
Recently, many works have been inspired by the success of deep learning in computer vision for plant diseases classification. Unfortunately, these end-to-end deep classifiers lack transparency which can limit their adoption in practice. In this paper, we propose a new trainable visualization method for plant diseases classification based on a Convolutional Neural Network (CNN) architecture composed of two deep classifiers. The first one is named Teacher and the second one Student. This architecture leverages the multitask learning to train the Teacher and the Student jointly. Then, the communicated representation between the Teacher and the Student is used as a proxy to visualize the most important image regions for classification. This new architecture produces sharper visualization than the existing methods in plant diseases context. All experiments are achieved on PlantVillage dataset that contains 54306 plant images.
- 1 April 2019
- Published by Elsevier in Future Generation Computer Systems
- Vol. 93, 176-187
Optimizing the cost of DBaaS object placement in hybrid storage systems
- 1 April 2019
- Published by Elsevier in Future Generation Computer Systems
- Vol. 93, 176-187
Optimizing the cost of DBaaS object placement in hybrid storage systems
- 13 December 2018
- Published by Springer Nature in Cluster Computing
- Vol. 22 (3) , 783-803
Designing data cubes in OLAP systems: a decision makers’ requirements-based approach
- 13 December 2018
- Published by Springer Nature in Cluster Computing
- Vol. 22 (3) , 783-803
Designing data cubes in OLAP systems: a decision makers’ requirements-based approach
- 5 October 2018
- Published by Taylor & Francis in Applied Artificial Intelligence
- Vol. 32 (9-10) , 907-923
Bi-Objective Optimization Method for Horizontal Fragmentation Problem in Relational Data Warehouses as a Linear Programming Problem
In this work, we relied on a particular exact method to solve NP-Hard problem of determining a horizontal fragmentation scheme in relational data warehouses. The method used is that of linear programming which is distinguished by other methods by the existence of practical methods that facilitate the resolution of problems that may be described in linear form. We quote the Simplex method and the interior points. To meet the linearity of the objective function and constraints, we used initially “De Morgan” theorem, which is based on properties of sets to transform and optimize decision queries, from any form to a linear one. In addition to designing and solving the selection problem of horizontal fragmentation technique, we considered the problem in two simultaneous objectives, namely: the number of Inputs/Outputs needed to run the global workload, and number of fragments generated to identify the best solutions compared to the concept of Pareto dominance. In addition, to carry out our work, we used the Benchmark APB1 invoked by a workload, to achieve satisfactory results.
- 5 October 2018
- Published by Taylor & Francis in Applied Artificial Intelligence
- Vol. 32 (9-10) , 907-923
Bi-Objective Optimization Method for Horizontal Fragmentation Problem in Relational Data Warehouses as a Linear Programming Problem
In this work, we relied on a particular exact method to solve NP-Hard problem of determining a horizontal fragmentation scheme in relational data warehouses. The method used is that of linear programming which is distinguished by other methods by the existence of practical methods that facilitate the resolution of problems that may be described in linear form. We quote the Simplex method and the interior points. To meet the linearity of the objective function and constraints, we used initially “De Morgan” theorem, which is based on properties of sets to transform and optimize decision queries, from any form to a linear one. In addition to designing and solving the selection problem of horizontal fragmentation technique, we considered the problem in two simultaneous objectives, namely: the number of Inputs/Outputs needed to run the global workload, and number of fragments generated to identify the best solutions compared to the concept of Pareto dominance. In addition, to carry out our work, we used the Benchmark APB1 invoked by a workload, to achieve satisfactory results.