Microblogging Short Text Classification Based on Word2Vec

Abstract

For the sparse features of the microblogging text, the author proposes a method of microblogging text classification based on the features extension by Word2Vec. We train the text by using Word2Vec tool and find the words which are similar to original features semantic as the features of short text. Then we expand the features to the original text, and finally classify the subject of microblogging text by using SVM method. Experimental results show that the method has high accuracy recall and F1 values compared with the traditional method of vector space model and LDA topic model.

Related Papers

• → Topic Modeling on News Articles using Latent Dirichlet Allocation(2022)10 cited
• → Hot topic extraction based on Chinese Microblog's Features topic model(2016)6 cited
• → Making Recommendations on Microblogs through Topic Modeling(2014)4 cited
• → Latent Dirichlet Allocation (LDA) and Topic modeling: models, applications, a survey(2017)164 cited
• → Topic Modelling of Swedish Newspaper Articles about Coronavirus: a Case Study using Latent Dirichlet Allocation Method(2023)2 cited