gOCCF: Graph-Theoretic One-Class Collaborative Filtering Based on Uninteresting Items

• DOI: 10.1609/aaai.v32i1.11707
• Publication Date: 2022-06-24T21:08:34Z
• Authors: Yeon-Chang Lee; Sang-Wook Kim; Dongwon Lee
• Keywords: Dense graph; Implementation

We investigate how to address the shortcomings of popular One-Class Collaborative Filtering (OCCF) methods in handling challenging “sparse” dataset one-class setting (e.g., clicked or bookmarked), and propose a novel graph-theoretic OCCF approach, named as gOCCF, by exploiting both positive preferences (derived from rated items) well negative unrated items). In capturing bipartite graph, further, we apply graph shattering theory determine right amount use. Then, develop suite graph-based based on random walk with restart belief propagation methods. Through extensive experiments using 3 real-life datasets, show that our gOCCF effectively addresses sparsity challenge significantly outperforms all 8 competing accuracy very sparse datasets while providing comparable best performing less datasets. The implementations used empirical validation are available for access: https://goo.gl/sfiawn.