Towards a hybrid approach of Primitive Cognitive Network Process and Agglomerative Hierarchical Clustering for music recommendation

Chun Guan, Kevin Kam Fung Yuen

Abstract: Clustering algorithms have been used in many real world applications including recommendation systems. This paper proposes PCNP-AHC, which is a hybrid approach of Primitive Cognitive Network Process (PCNP) and Agglomerative Hierarchical Clustering (AHC) to cluster music pieces on the basis of user's preferences and similarities between music pieces. PCNP is an ideal alternative of Analytic Hierarchy Process (AHP) to quantify weights of attributes which are used in clustering process. The application of PCNP-AHC for music recommendation is demonstrated.