Short Text Classification Based on Keywords Extension

摘要

Due to the number of short texts in news is small, the traditional text processing method often causes the lack of semantic information when analyzing the news text, which becomes one of the bottlenecks that restrict the performance of short text classification. This paper uses the external corpus to train the Word2Vec model, expands the keywords extracted by the traditional keyword extraction algorithm based on external semantic information, and studies the feasibility of extending short text keywords based on external semantic information according to different extension methods. Finally, KNN (K-Nearest Neighbor) algorithm is used to verify that the proposed method improves the performance of short text classification in news compared with the classical algorithm, which is close to the current mainstream text classification algorithm.