Research on talent data analysis method based on Text Mining

Abstract

In order to solve the problem of low clustering contour coefficient caused by inaccurate keyword extraction of talent data information, a talent data analysis method based on text mining is proposed. Through the preprocessing of word segmentation and stop words, the text set of talent data is established, and the keyword graph is constructed by text mining, and the information keywords are obtained according to the weight iteration results. The keywords in this paper are transformed into the form of multi-dimensional vector, and the similarity is calculated to get the results of text analysis. The experimental results show that the contour coefficient of the proposed method is 0.736, which is 0.267 and 0.221 higher than that of the K-means and single pass methods, respectively. The design method of this paper has a reasonable clustering performance, which is suitable for talent big data analysis.