Semi-supervised text clustering method and device fusing pairwise constraints and keywords

Abstract

The invention discloses a semi-supervised text clustering method and device fusing pairwise constraints and keywords. The method comprises the following steps of: fusing pairwise constraints to assist in text clustering to obtain an initial feature word weight; fusing the pairwise constraints and keywords and performing the semi-supervised clustering at the same time based on the obtained initial feature word weight; and evaluating and selecting a clustering result according to a user satisfaction degree. The device provided by the invention comprises a pre-processing module, a text clustering module fusing pairwise constraints, a semi-supervised text clustering module fusing pairwise constraints and the keywords, and an evaluation and selecting result module. Since the semi-supervised text clustering method provided by the invention continuously adds keyword information on the basis of fusing pairwise constraint information, the keyword information is used for adjusting the corresponding feature word weight while applying the pairwise constraints to learning the feature word weight; and therefore, the two prior information can be mutually influenced and promoted to obtain a more accurate clustering result.

Description

technical field [0001] The present invention relates to a text clustering method, in particular to a semi-supervised text clustering method and device for fusing pairwise constraints and keywords. Background technique [0002] Traditional text clustering usually adopts an unsupervised learning mechanism, which automatically divides texts with similar topics together and divides texts with different topics. However, the performance of such text clustering methods is often unsatisfactory. This is caused by many reasons, such as inability to interact with the user, segmentation results are difficult to understand, and so on. [0003] In recent years, many researchers have adopted semi-supervised learning strategies to integrate prior information to assist text clustering, which has effectively improved the quality of clustering and the understandability of segmentation results. Prior information mainly includes instance layer information and attribute layer information. Inst...

AI Technical Summary

Problems solved by technology

Although the semi-supervised text clustering methods based solely on one of the restrictive information can effectively improve the clustering quality, they do not comprehensively consider the fusion of these two different types of information when processing, resulting in the accuracy of the clustering results and the Not very stable

Images