Sentiment user profile analysis based on forgetting curve in mobile environments

Abstract

The user profile is a way of describing user characteristics using collected information about what the user prefers. However, it cannot be used for other applications with a general purpose, because it is generally created for personal recommendation and search. Furthermore, there are restrictions to analyze the user profile in terms of temporal influence in mobile environments. In this paper, in order to overcome these limitations, we propose a sentiment user profile analysis, which can analyze individual user characteristics based on the forgetting curve in mobile environments. In order to evaluate the effectiveness and feasibility of the proposed analysis, it has been implemented and applied to clothing domain. Experimental results show that the proposed analysis extracts the individual user characteristics with better accuracy than the existing method.