Improvement of LDA Topic Mining Algorithm and Its Application in Short Text
Kai Li, Chunmei Li · Journal of Physics Conference Series · 2019
In order to quickly provide a large number of short text themes, an improved linear discriminant analysis (LDA) topic mining algorithm is proposed in this paper. Firstly, the acquisition method of obtaining the traditional short text themes is first analyzed. Focusing on the shortcomings of the original algorithm, the improvement process of the improved topic mining algorithm based on LDA is introduced. Finally, through the mining of short text themes by college students as an example, the improved LDA algorithm is improved in the accuracy of short text processing.