Research on archives text classification based on Naive bayes

摘要

This paper analyzes the data resources of archives in Gansu Province by combining with the characteristics of archives resources, and combines with Naive Bayesian classification algorithm to realize the application of archives resource classification. According to the characteristics of the file data, select the attribute that matches the text of the file text, and use the TFIDF algorithm in the file text feature attribute selection. The experimental results show that the classification model is suitable for the classification of archival text resources, and the function of automatic classification of archives is realized. Compared with the traditional Naive Bayesian classification method, the classification model proposed in this paper is 1% -2% for the classification efficiency of archives, it is a more effective classification model for the archives.