Nowadays, Swarm Intelligence (SI) is considered as one of the most promising class of soft computing and bio-inspired algorithms for solving complex optimization problems. Particle Swarm Optimization (PSO) is by far the most popular and widely used swarm intelligence algorithms because of its efficient simplicity, high flexibility, and fast convergence speed. On the other hand, Information Retrieval (IR) systems are still suffering from shortcomings related to query items rating and thus returning irrelevant answers. Therefore, in this paper, we adopt PSO algorithm to re-rate query items and enhance the retrieval effectiveness of IR systems. The key idea is to reduce the noise effect of widespread items by minimizing the number of answers that contain these frequent items, and increase the discriminative effect of rare items by maximizing the number of answers that contain these infrequent items. Extensive experiments have been conducted using real data from an online medical dataset. The results show that the proposed algorithm improves the retrieval effectiveness and demonstrates a substantial enhancement over the state-of-the-art.
Khennak, I., Drias, H., Drias, Y. (2021). Particle Swarm Optimization for Query Items Re-rating. In Hybrid Intelligent Systems 20th International Conference on Hybrid Intelligent Systems (HIS 2020), December 14-16, 2020 (pp.729-739). Springer Science and Business Media Deutschland GmbH [10.1007/978-3-030-73050-5_71].
Particle Swarm Optimization for Query Items Re-rating
Drias Y.
2021
Abstract
Nowadays, Swarm Intelligence (SI) is considered as one of the most promising class of soft computing and bio-inspired algorithms for solving complex optimization problems. Particle Swarm Optimization (PSO) is by far the most popular and widely used swarm intelligence algorithms because of its efficient simplicity, high flexibility, and fast convergence speed. On the other hand, Information Retrieval (IR) systems are still suffering from shortcomings related to query items rating and thus returning irrelevant answers. Therefore, in this paper, we adopt PSO algorithm to re-rate query items and enhance the retrieval effectiveness of IR systems. The key idea is to reduce the noise effect of widespread items by minimizing the number of answers that contain these frequent items, and increase the discriminative effect of rare items by maximizing the number of answers that contain these infrequent items. Extensive experiments have been conducted using real data from an online medical dataset. The results show that the proposed algorithm improves the retrieval effectiveness and demonstrates a substantial enhancement over the state-of-the-art.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
