Application of Computer Data Mining Technology in the Analysis of Emotional Law

Abstract

In recent years, with the in-depth development of the Internet, computer data mining technology has gradually matured, social networks have been integrated into people’s daily lives, and more and more Internet users have expressed their personal emotions and commented on social hotspots on social networks. In this regard, social network public opinion analysis plays a vital role, and has attracted widespread attention from academia and industry. In order to better excavate Weibo users’ opinions and ideas on social events, and track the development of the events, it is more conducive to the work of public opinion monitoring and rumor control. This paper expounds the research status of data mining and the current situation of psychological problems of college students. It introduces computer data mining technology into the analysis of college students’ emotional regularity, and introduces the concepts, functions, technologies, methods and processes of data mining. This paper proposes a clustering method based on the characteristics of the semantic recessive sentiment of frequent itemsets, which fully considers the implicit sentiment regularity of semantics in Weibo. Firstly, we define the frequent feature word set of Weibo based on the characteristics of dominant sentiment, and use the maximum frequent items. Clusters to obtain initial clusters of regular emotions; for text overlap between initial clusters, an inter-cluster overlap reduction algorithm based on short text extended semantic membership is proposed to obtain completely separated initial clusters; based on cluster semantic similarity matrix, agglomerated emotions are given Regular clustering method. Finally, the validity of the text method is verified by the training corpus data provided by the NLP & CC2013 evaluation. It can be known from the test results that when the minimum support degree θ of the cluster is 0.5-0.6, the clustering effect of the analysis of sentiment regularity is better.