Item recommendation using user feedback data and item profile

Matrix factorization (MS) is a collaborative filtering (CF) based approach, which widely used for recommendation systems (RS). In this research work, we deal with the content problem users in management system (CMS) on users' feedback data. The CMS applied publishing and pushing curated to employees of company or an organization. Here, have user's data solve problem. We prepare individual user profiles then generate results different categories, including Direct Interaction, Social Share, Reading Statistics, Subsequently, analyze effect categories results. shown that impacts accuracy. best performance achieves if include all types task. also incorporate similarity as regularization term into MF model designing hybrid model. Experimental proposed demonstrates better compared traditional MF-based models.