Research On The Application Of Text Mining Technology In Students' Teaching Evaluation

In 2018,the Ministry of Education issued the first national standard for teaching quality in the field of higher education in China.This measure can show that China attaches great importance to improving teaching quality.As a crucial means of teaching quality evaluation in Colleges and universities,how to effectively identify,refine and analyze students' concerns and emotions from a large number of student teaching evaluation texts,so as to realize the purpose of school managers to deeply understand the relevant information of curriculum teaching,so as to improve the teaching quality,which is the focus of this paper.With the help of text mining technology,this paper analyzes and studies a large number of comment data in students' teaching evaluation,constructs a dictionary in the field of students' teaching evaluation,and mines students' emotional tendency from students' comments.The main work of this paper is as follows:1.Firstly,we obtain and sort out the text data of students' teaching evaluation in Y University,and then use Python language to write code in Anaconda to complete the preprocessing of students' comments,Jieba word segmentation,statistical analysis and visual presentation of word cloud.2.In view of the low recognition rate of new words in students' comments by the basic emotion dictionary,this paper uses a fusion method of so-pmi and word2 vec to construct the emotion dictionary in the field of students' teaching evaluation based on the basic emotion,and evaluates its emotion analysis performance.At the same time,based on the dictionary in the field of students' evaluation of teaching,this paper analyzes the emotional tendency of students' evaluation.3.The LDA model is used to mine the topics of student comments,and the results of topic classification are visualized with the help of pyldavis.The results show that the analysis performance of the dictionary in the field of student evaluation constructed by S-W method is better than that extended by so-pmi and word2 vec method to a certain extent.In derogatory comments,the f eigenvalue is increased by 4.9% and 4.3% respectively.Based on the constructed dictionary and LDA model in the field of students' teaching evaluation,this paper analyzes students' comments in many aspects,and realizes the multi-dimensional feedback of teachers' classroom teaching quality,which has good application value and practical significance.