Nesrine Ben Yahia

Towards an Intelligent Decision Making Support

This paper presents an intelligent framework that combines case-based reasoning (CBR), fuzzy logic and particle swarm optimization (PSO) to build an intelligent decision support model. CBR is a useful technique to support decision making (DM) by learning from past experiences. It solves a new problem by retrieving, reusing, and adapting past solutions to old problems that are closely similar to the current problem. In this paper, we combine fuzzy logic with case-based reasoning to identify useful cases that can support the DM. At the beginning, a fuzzy CBR based on both problems and actors’ similarities is advanced to measure usefulness of past cases. Then, we rely on a meta-heuristic optimization technique i.e. Particle Swarm Optimization to adjust optimally the parameters of the inputs and outputs fuzzy membership functions.

Towards an Intelligent Application of Large Scale Community Detection to Support Collaboration During Emergency Management

During crisis situations, people often use social media to seek for help and to find new collaborators who can help them in emergency management. In this context, we propose an intelligent application to find and recommend potential and relevant collaborators through social media. This application is based on a large scale contextualized community detection to compose dynamic groups. To do so, we propose to reuse a new community detection algorithm that considers simultaneously the network structure (social connections) and profiles homophily (similarities). An application of the proposed solution and a comparison with another community detection algorithm evaluates its performance.

Evaluating community detection using a bi-objective optimization

Community detection consists on a partitioning networks technique into clusters (communities) with weak coupling (external connectivity) and high cohesion (internal connectivity). In order to measure the performance of the clustering, the network modularity is largely used, a metric that presents the cohesion and the coupling of communities. In this paper, a global and bi-objective function is proposed to evaluate community detection. This function combines modularity (based on structure and edges weights) and the inter-classes inertia (based on nodes weights). Then, we rely on a computational optimization technique i.e. Particle Swarm Optimization to maximize this bi-objective quality. Finally, a case study evaluates the proposed solution and illustrates practical uses.

Hybrid modeling approach for contextualized community detection in multilayer social network: emergency management case study

Abstract Within multilayer social networks, finding relevant communities for each specific situation has been a challenging task. Thus, modeling these social networks is the key issue for the process of contextualized community detection. However, traditional formalisms for representing multilayer social networks suffer from the lack of semantics. In the scope of this paper, we propose a hybrid modeling approach to represent participants and community detection context in multilayer social network. This approach combines a semantically rich description of social data (Ontology-based model) with a powerful mathematical abstraction (Graph-based model). Furthermore, we present a modeling scenario in the field of emergency management to illustrate how the proposed model can be used to contextualize community detection within a real social network. Finally, a comparison with another modeling approach is given in order to evaluate the proposed model performance.

Towards Contextualizing Community Detection in Dynamic Social Networks

With the growing number of users and the huge amount of information in dynamic social networks, contextualizing community detection has been a challenging task. Thus, modeling these social networks is a key issue for the process of contextualized community detection. In this work, we propose a temporal multiplex information graph-based model to represent dynamic social networks: we consider simultaneously the social network dynamicity, its structure (different social connections) and various members’ profiles so as to calculate similarities between “nodes” in each specific context. Finally a comparative study on a real social network shows the efficiency of our approach and illustrates practical uses.

Hybrid community detection approach in multilayer social network: Scientific collaboration recommendation case study

Within real-world social networks people are linked with multiple types of relationships, which brings new challenges in community detection for multilayer social network where each layer represents one type of relationships. However, most of existing approaches consist on transforming the problem into a classical problem of community detection in monoplex network. In this work, we propose a new hybrid community detection approach in multilayer social networks. This approach considers simultaneously the network structure (different social connections) and the homophily of participants (similarities between users). To do so we propose a new multiplex information graph model to represent multilayer social network. Then, we adapt a combined community detection algorithm to the multiplex case. Furthermore, an example in the field of scientific collaboration recommendation is given to illustrate the practical usefulness of the proposed approach. Finally, a comparison with other community detection approaches evaluates its performance.