Multi-criteria Recommender Systems Based on Multi-objective Hydrologic Cycle Optimization

• Published in: Advances in Swarm Intelligence Vol. 11656; pp. 92 - 102
• Main Authors: Geng, Shuang, Zhang, Churong, Yang, Xuesen, Niu, Ben
• Format: Book Chapter
• Language: English
• Published: Switzerland Springer International Publishing AG 2019; Springer International Publishing
• Series: Lecture Notes in Computer Science
• Subjects: Hydrologic cycle optimization; Multi-objective problem; Pareto optimal solutions; Recommender systems
• ISBN: 3030263533; 9783030263539
• ISSN: 0302-9743; 1611-3349
• DOI: 10.1007/978-3-030-26354-6_9

Traditional recommendation systems always consider precision as the unique evaluation standard. However, diversity and user tendency are also important for recommendation system performance. The implementation of multiple performance factors can be expressed as a multi-objective optimization problem (MOP). This paper attempts to combine multi-objective optimization algorithm with recommendation algorithm to solve this multi-objective recommendation problem. A novel multi-objective heuristic algorithm called Multi-objective Hydrologic Cycle Optimization (MOHCO) is proposed. MOHCO simulates the water flow, infiltration, evaporation and precipitation processes in nature, and aims to find a set of Pareto optimal solutions. Experimental tests on Grouplens – MovieLens 100K movie recommendation dataset demonstrate that MOHCO outperforms other heuristic algorithms including MOEAD, NSGAII, NSGAIII, MOPSO.