Research on Emotional Word Clustering Based on Affinity Propagation

摘要

In this paper, the affinity propagation algorithm algorithm(AP) and its related parameters are studied, and the volatility is used to measure the intensity of the data concussion. The two parameters of preference and damping factor are studied. Then, according to the need of clustering analysis of emotion words, the AP algorithm is improved and applied to the generation of emotional word clustering. From the experimental results, the improved algorithm has better time performance and emotional word clustering performance.