Abir Betka

Abir Betka, Zeyd Ferhat, Riyadh Barka, Selma Boutiba, Zineddine Kahhoul, Tiar Lakhdar, Ahmed Abdelali, Habiba Dahmani. Proceedings of ArabicNLP 2023. 2023.

Optimization is an essential tool that improves the problem’s results and leads to the best possible solution in different research domains. In data mining, optimization can be used to formulate a suitable design for the learning model, in order to enhance the decision-making step. Data mining is considered as a promising approach that exploits large volume of data and transforms them into helpful information for an appropriate decision. The diverse use of data mining has shown its needs to a training process, neural networks have been used successfully in this context. They have been also combined with metaheuristic techniques to provide superior results. Symbiotic Organisms Search (SOS) is one of the most important and powerful metaheuristics due to its simplicity and robustness. However, it suffers from the premature convergence. To alleviate this shortcoming, we propose an improved version of the SOS called VSOS (Velocity Symbiotic Organism Search), in which a new term, velocity, is introduced. This term is involved to realise a balance between local search (exploit) and global search (explore). The proposed method is firstly benchmarked on 22 test functions. The VSOS method is then combined with Multi-Layer Perceptron Artificial Neural Network (MLP-ANN) for classification in data mining process. Three popular datasets are employed for this experiment, and the results, in term of classification accuracy, are compared with those obtained with other metaheuristics, such as BBO, GA, PBIL, PSO, HWBA, SOS based on clustering process and the conventional SOS. Experimental results indicate that the proposed approach can provide classification accuracy up to 99% with some datasets, outperforming the other cited methods, which confirms its high performance and credibility of treatment.

This paper is concerned with the stability and stabilization problem of a Takagi-Sugeno fuzzy (TSF) system. Using a non-quadratic function (well-known integral Lyapunov fuzzy candidate (ILF)) and some lemmas, new sufficient conditions are established as linear matrix inequalities (LMIs), which are solved with a stochastic fractal search (SFS). The main advantage of the technique used is its small conservatives. Motivated by the mean value theorem, a state feedback controller based on a non-quadratic Lyapunov function is designed. Unlike other approaches based on poly-quadratic Lyapunov candidates, stability conditions of the closed loop are obtained in LMI regions. It is important to highlight that the time derivatives of membership functions do not appear in the used line integral Lyapunov function, which is the well-known problem of poly-quadratic Lyapunov functions. A numerical example is given to show the advantages and the utility of the integral Lyapunov fuzzy candidate, which provides a wider feasibility region than other Lyapunov functions.