Probabilistic Models Based on Non-Negative Matrix Factorization for Inconsistent Resolution Dataset Analysis

DOI

Bibliographic Information

Other Title
  • 異粒度データ分析のための非負値行列分解に基づく確率モデル

Search this article

Description

In this paper, we tackle with the problem of analyzing datasets with different resolution such as a pair of user's individual data and user group's data, for example “userA visited shopA 5 times” and “users whose attributes are men purchased itemA 80 times in total”. In order to establish a basic approach to this problem, we focus on the simplified scenario and propose a new probabilistic model called probabilistic non-negative inconsistent-resolution matrices factorization (pNimf). pNimf is rigorously derived from the data generative process using latent high-resolution data which underlie low-resolution data. We conduct experiments on real purchase log data and confirm that the proposed model provides superior performance, and that the performance improves as the number of low-resolution data increases. Moreover, by deriving an extended method based on the proposed model, we show that the proposed model can become the basic approach to various problems of analyzing dataset with different resolution.

Journal

Keywords

Details 詳細情報について

Report a problem

Back to top