M. MEHENNI Tahar

MCA

Directory of teachers

Department

Informatics Department

Research Interests

Data Mining and Machine Learning Bioinformatics Software engineering Metaheuristics Information Systems Data Science

Contact Info

University of M'Sila, Algeria

Email : tahar.mehenni@univ-msila.dz

On the Web:

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

Recent Publications

2024-11-19

Leveraging machine learning models for food security and crisis management in Algeria

The food sector in Algeria and many countries around the world faces multiple challenges, ranging from environmental pollution and climate change to fluctuations in production and distribution processes, increasing the complexities of maintaining food safety and mitigating health and environmental risks. This paper addressed analysis and prediction of food insecurity risks through the development of predictive models. Various models were reviewed and applied to real market data, focusing on predicting product prices and assessing food safety risks. The results showed that Polynomial Regression and Decision Tree models were the most efficient in accurately predicting and classifying the products, confirming their effectiveness in food safety applications. These findings contribute to providing accurate recommendations and making informed decisions to enhance food safety and better crisis management in the future.
Citation

M. MEHENNI Tahar, Hadil Bouti, Fatima Zohra Zerrouak, , (2024-11-19), "Leveraging machine learning models for food security and crisis management in Algeria", [international] International Conference on Information and Communication Technologies for Disaster Management (ICT-DM) , Setif-Algeria

Read Full Paper
2024-11-10

Empowering Customer Service with AI: A Multilingual Chatbot for Mobile Companies

Nowadays, people are using more and more advanced technology in communication. Consequently, the brand-customer relationship has never been more intense, thanks to rapid technological advancements. Customers are preferring self-service options satisfying autonomy and enabling them to make purchases or access information without engaging with human representatives. Conversational systems (chatbots) present the most interesting solution to both enhancing customerbrand communication and improving the customer experience. Mobile companies are considering this technology in order to perform a better engagement with clients using software platforms, that can offer regular chat functionalities, i.e. giving more information, solving problems, making appointments, and more other advanced features. Moreover, with the presence of global customer bases, mobile operators need conversational systems that can handle queries using multiple languages. Therefore, implementing chatbots seems to be an effective and profitable solution for both parties. These systems have to be designed to simulate call center employees by responding to common inquiries and guiding customers in their languages, in order to reduce the long time spent searching for information. In this paper, we present a rule-based chatbot for mobile company customers that supports three languages: Arabic, English and French, and automatically recognizes the user's language without any additional information. The developed prototype is evaluated via a pre-defined questionnaire and the results are promising, indicating that this solution could be advantageous for mobile companies and other organizations.
Citation

M. MEHENNI Tahar, Samiha Attaoua, Chaima Guesmia, , (2024-11-10), "Empowering Customer Service with AI: A Multilingual Chatbot for Mobile Companies", [international] THE SIXTH INTERNATIONAL SYMPOSIUM ON INFORMATICS AND ITS APPLICATIONS ISIA 2024 , University of M'sila

Read Full Paper
2024-10-22

L'IA omniprésente : Innovations dans les secteurs primaire, tertiaire et mobile

L'intelligence artificielle (IA) est devenue une force transformatrice dans notre société moderne, révolutionnant la manière dont nous vivons, travaillons et interagissons avec le monde qui nous entoure. Définie comme la capacité des systèmes informatiques à  imiter l'intelligence humaine, l'IA englobe un large éventail de technologies, dont l'apprentissage automatique et les réseaux neuronaux. Au cours de la dernière décennie, l’IA s’est intégrée à divers secteurs d’actitivité des êtres humains. Cette période a connu une augmentation spectaculaire des outils, des applications et des plateformes basées sur l’IA et le machine learning (ML). Ces technologies ont eu un impact sur les soins de santé, la fabrication, le droit, la finance, la vente au détail, l’immobilier, la comptabilité, le marketing numérique et plusieurs autres domaines. Aujourd’hui, l’IA est un moteur d’innovation dans des secteurs variés, du secteur primaire aux services mobiles en passant par le secteur tertiaire.
Cet article explore certaines de ces innovations et comment elles façonnent notre quotidien. Des exemples de PFE réalisés par des étudiants en Master sont présentés.
Citation

M. MEHENNI Tahar, (2024-10-22), "L'IA omniprésente : Innovations dans les secteurs primaire, tertiaire et mobile", [national] Premier Colloque sur les pratiques de l’intelligence artificielle , Bordj Bou Arriridj, Algeria

Read Full Paper
2024-05-07

A multilingual chatbot for supporting mobile companies complaints

People want to communicate with technology in the same manner they communicate with other human beings, and the communication between brands and their clients has never been so intense as it is nowadays. With the rapid development of technology, the customer experience is changing dramatically. Customers want more autonomy and self-service options, preferring to make a purchase or get information without interacting with the human representative of the brand. Therefore, the use of chatbots in customer service can be a solution to the crucial issue of improving customer-brand communication. Companies are using this technology to create better engagement with their clients with the help of messaging platforms, to offer a regular chat function, inmessage purchasing, and many other advanced functions. Consequently, an effective and profitable solution to this problem for two parties is a chatbot. It is a software application that simulates a call center employee that
responds to common questions and directs customers to reduce the time spent by the user in finding the right information. In our work, we explored two different chatbot systems, the first one is a chatbot that has has been designed and trained in Google cloud platform (Dialogflow) and the second is a rule-based chatbot that supports three languages and automatically recognizes the user's language among the languages Arabic is not supported by the first type.
Citation

M. MEHENNI Tahar, (2024-05-07), "A multilingual chatbot for supporting mobile companies complaints", [international] the 3rd European Computing , Greece

Read Full Paper
2024-04-18

IMPLICATIONS OF DIGITAL TECHNOLOGIES FOR TRANSPARENCY AND ACCOUNTABILITY

The topic of this paper is focused on the third point: transparency, accountability, and anti-corruption. The main research question can be formulated as follows: Do ICTs in e-Government encourage transparency, promote accountability and reinforce anti-corruption measures in governments? This paper presents an analysis of the role that e-Government has played during the last decade in order to promote transparency, accountability, and anti-corruption measures. It starts with a definition of the principal concepts of the study. It then presents the links between transparency, accountability and empowerment. Effects of ICTs on corruption are discussed before presenting ICTs for reporting with some examples of digital complaint mechanisms in some countries, where Narakom platform of Algeria is introduced.
Citation

M. MEHENNI Tahar, (2024-04-18), "IMPLICATIONS OF DIGITAL TECHNOLOGIES FOR TRANSPARENCY AND ACCOUNTABILITY", [national] الملتقى الوطني حول دور الرقمنة في الوقاية من الفساد ومكافحته , Jijel, Algeria

Read Full Paper
2022

Analysis of road accident factors using Decision Tree Algorithm: a case of study Algeria

Road accidents become a worldwide health issue. With the enormous number of death and injuries, this problem pushes governments to create solutions to reduce those statistics. One of the solving ways is using machine learning algorithms, and with the data collected from road accidents, we can increase traffic safety. In this research, we use a decision tree model to analyze road accidents that happened in Algeria. Then, we do a comparison with some similar works using accuracy as a performance evaluation metric. This work can help government and traffic safety entities to improve road safety and minimize the number of accidents, also, it can help other researchers to develop other models in the analysis of traffic accidents in Algeria and other countries.
Citation

M. MEHENNI Tahar, (2022), "Analysis of road accident factors using Decision Tree Algorithm: a case of study Algeria", [international] IEEE , Algeria

Read Full Paper
2020

Multi-database Mining: A Framework for Heterogeneous Relational Databases

Multiple heterogeneous databases are widely used in many disciplines, such as decision support and medical research. Therefore new data mining approaches are required handling databases coming from multi-database systems which are relational and generally heterogeneous. In this paper, we propose a framework for classification task over multiple relational heterogeneous databases. The proposed framework is composed of two components. The first one resolves the heterogeneity problem using schema matching techniques. The second is a decision tree classification approach using a strategic communication technique based on utility theory. Experiments performed on real databases were very satisfactory and showed that the proposed framework is giving an efficient solution for the multi-database mining.
Citation

M. MEHENNI Tahar, (2020), "Multi-database Mining: A Framework for Heterogeneous Relational Databases", [international] ISPR , Tunisia

Read Full Paper
2019

Identifying Natural Road Accidents factors using Spatial Big Data Mining Techniques

Identifying Natural Road Accidents factors using Spatial Big Data Mining Techniques
Citation

M. MEHENNI Tahar, Belkorchia Mouaouia, , (2019), "Identifying Natural Road Accidents factors using Spatial Big Data Mining Techniques", [international] International Conference on Data Science and Applications , Turkey

Read Full Paper

A new deep learning-based chatbot system for the Customer service companies

People want to communicate with technology in the same manner they communicate with other human beings, and the communication between brands and their clients has never been so intense as it is nowadays. With the rapid development of technology, the customer experience is changing dramatically. Customers want more autonomy and self-service options, preferring to make a purchase or get information without interacting with the human representative of the brand. Therefore, the use of chatbots in customer service can be a solution to the crucial issue of improving customer-brand communication. Companies are using this technology to create better engagement with their clients with the help of messaging platforms, to offer a regular chat function, in-message purchasing, and many other advanced functions. In our work, we have explored two deferent chatbot systems, the first bot is an open domain deep learning chatbot that has been trained on our personal computer, and the second one is a customer service chatbot that designed and trained in Google cloud platform.
Citation

M. MEHENNI Tahar, (2019), "A new deep learning-based chatbot system for the Customer service companies", [national] Revue de l'Information Scientifique et Technique , CERIST , Algeria

Read Full Paper

SCATTER: Fully Automated Classification System across Multiple Databases

Data mining approaches performed recently use data coming from a single table and are not adapted to multiple tables. Moreover, computer network expansion and data sources diversity require new data mining systems handling databases heterogeneity in multi-database systems. In this paper, we propose SCATTER: a fully automated classification system from multiple heterogeneous databases. SCATTER is composed of three components. The first component uses schema matching techniques to find foreign-key links across the multi-database system. The second component tries to find the most useful links that are critical for producing accurate classes across multiple databases. The last component is a decision tree classification algorithm which exploits the useful links discovered automatically across the databases. Experiments performed on real databases were very satisfactory with an average accuracy of 86.5% and showed that SCATTER system succeeded in achieving a fully automated classification from multiple heterogeneous databases.
Citation

M. MEHENNI Tahar, (2019), "SCATTER: Fully Automated Classification System across Multiple Databases", [national] International Journal of Computing and Digital Systems , University of Bahrain

Read Full Paper
2018

Integrating phylogeny-based tabu search in multiple sequence alignement

Nowadays, Multiple Sequence Alignment (MSA) is a central problem in computational biology for discovering functional, structural, and evolutionary information of biological sequences. However, MSA problem is becoming more and more difficult when handling sequences having low similarities. Therefore, recent MSA approaches do not always provide consistent solutions. Tabu Search is a very useful and widely applied meta-heuristic approach in solving real world optimization problems from many domains. For the alignment of multiple sequences, which is a NP-hard problem, we apply a tabu search algorithm improved by several neighborhood generation techniques using phylogeny concepts guide trees. The algorithm is tested with the BAliBASE benchmarking database, and experiments showed encouraging results compared to the recents algorithms.
Citation

M. MEHENNI Tahar, (2018), "Integrating phylogeny-based tabu search in multiple sequence alignement", [international] International conference on applied mathematics, modeling and Life Sciences , Turkey

Read Full Paper
2017

Handbook of research on geographic information systems applications and advancements

Voluminous geographic data have been, and continue to be, collected from various Geographic Information Systems (GIS) applications such as Global Positioning Systems (GPS) and high-resolution remote sensing. For these applications, huge amount of data is maintained in multiple disparate databases and different in spatial data type, file formats, data schema, access mechanism, etc. Spatial data mining and knowledge discovery has emerged as an active research field that focuses on the development of theory, methodology, and practice for the extraction of useful information and knowledge from massive and complex spatial databases. This chapter highlights recent theoretical and applied research in geographic knowledge discovery and spatial data mining in a distributed environment where spatial data are dispersed in multiple sites. The author will present in this chapter, an overall picture of how spatial …
Citation

M. MEHENNI Tahar, (2017), "Handbook of research on geographic information systems applications and advancements", [national] , IGI Global

Read Full Paper
2015

A tabu search using guide trees-based neighborhood for the multiple sequence alignment problem

Nowadays, current Multiple Sequence Alignment (MSA) approaches do not always provide consistent solutions. In fact, alignments become increasingly difficult when treating low similarity sequences. Tabu Search is a very useful metaheuristic approach in solving optimization problems. For the alignment of multiple sequences, which is a NP-hard problem, we apply a tabu search algorithm improved by several neighborhood generation techniques using guide trees. The algorithm is tested with the BAliBASE benchmarking database, and experiments showed encouraging results compared to the algorithms studied in this paper.
Citation

M. MEHENNI Tahar, (2015), "A tabu search using guide trees-based neighborhood for the multiple sequence alignment problem", [international] International Conference on Circuits, Systems, Communications and Computers , Greece

Read Full Paper

Multiple guide trees in a tabu search algorithm for the multiple sequence alignment problem

Nowadays, Multiple Sequence Alignment (MSA) approaches do not always provide consistent solutions. In fact, alignments become increasingly difficult when treating low similarity sequences. Tabu Search is a very useful meta-heuristic approach in solving optimization problems. For the alignment of multiple sequences, which is a NP-hard problem, we apply a tabu search algorithm improved by several neighborhood generation techniques using guide trees. The algorithm is tested with the BAliBASE benchmarking database, and experiments showed encouraging results compared to the algorithms studied in this paper.
Citation

M. MEHENNI Tahar, (2015), "Multiple guide trees in a tabu search algorithm for the multiple sequence alignment problem", [international] IFIP International Conference on Computer Science and its Applications , Algeria

Read Full Paper

Integration of useful links in distributed databases using decision tree classification

Nowadays, distributed relational databases constitute a large part of information storage handled by a variety of users. The knowledge extraction from these databases has been studied massively during this last decade. However, the problem still present in the distributed data mining process is the communication cost between the different parts of the database located naturally in remote sites. We present in this paper a decision tree classification approach with a low cost communication strategy using a set of the most useful inter-base links for the classification task. Different experiments conducted on real datasets showed a significant reduction in communication costs and an accuracy almost identical to some traditional approaches.
Citation

M. MEHENNI Tahar, (2015), "Integration of useful links in distributed databases using decision tree classification", [international] International Conference on Information Systems and Economic Intelligence , Tunisia

Read Full Paper
2013

Single Machine Scheduling Problem with non-renewable Resource

In this work, we address the single machine scheduling problem with non-renewable (consumable) resource constraints where the objective is to minimize the sum of weighted completion times. In this problem, a job can be processed if a sufficient quantity of resource is available. We developed two metaheuristics to resolve this problem. The first one is a local search method, which is based on a truncated branch-and bound algorithm to find the neighbourhood. Two lower bounds are proposed in order to evaluate the best one embedded in the branch-and-bound search. The second metaheuristic is a tabu search using a dynamic tabu list and a new diversification strategy, called k-first strategy, which is based on a list of the k first best solutions not chosen in the neighbourhood search. The tabu search is evaluated with two structures of the neighbourhood. Finally we present computational experiments in order to show the efficiency of the proposed metaheuristics.
Citation

M. MEHENNI Tahar, (2013), "Single Machine Scheduling Problem with non-renewable Resource", [international] ALGERIAN TURKISH INTERNATIONAL DAYS ON MATHEMATIC , Turkey

Read Full Paper
2012

Data mining from multiple heterogeneous relational databases using decision tree classification

Nowadays, the expansion of computer networks and the diversity of data sources require new data mining approaches in multi-database systems. We propose a classification approach across multiple heterogeneous relational databases. More specifically, given a set of inter-related databases, we use a regression model for predicting the most useful links that will be connected to build a multi-relational decision tree. Experiments performed on different real and synthetic databases were very satisfactory compared with previous classification approaches in multiple databases.
Citation

M. MEHENNI Tahar, Abdelouahab Moussaoui, , (2012), "Data mining from multiple heterogeneous relational databases using decision tree classification", [national] Pattern Reconginition Letters , Elsevier

Read Full Paper
2006

utilisation des métaheuristiques pour résoudre un problème d’ordonnancement sur machine a contrainte de ressource non renouvelable

Le problème qu’on s’est proposé d’étudier est l’ordonnancement d’un certain nombre de
tâches sur une machine unique. Chaque tâche nécessite une certaine quantité de ressource pour qu’elle
soit exécutée. Cette ressource est non renouvelable (ou consommable), c’est-à-dire elle diminue à
chaque utilisation par les tâches, jusqu’à une certaine limite, exemples : les lubrifiants, le carburant, la
matière première, l’énergie sur batterie, les produits semi-finis, …
Citation

M. MEHENNI Tahar, (2006), "utilisation des métaheuristiques pour résoudre un problème d’ordonnancement sur machine a contrainte de ressource non renouvelable", [national] University of M'sila

Read Full Paper
← Back to Researchers List