<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>As an important guide in the growth process of college students, how to communicate effectively with college students in discourse is a problem faced by college counselors. The LDA college counselor discourse text analysis model is created using Word2vec after processing the discourse of college counselors through text conversion in this paper. The model improves the topic resolution of the LDA model by removing the deactivated words in the discourse text and enriches the word vector data of the discourse text by combining it with the function of expanding feature words of Word2vec. The constructed model is applied to analyze the current situation of the tutor’s discourse themes and subjects by combining the topics generated during the discourse exchange between the tutor and the student and the corresponding generation probability and constructing the tutor discourse system in colleges and universities according to the analysis results. In the analysis of discourse topics, the generation probability of the topic word of personal emotion is the largest among the 20 topics, which is 0.502498, exceeding 0.5. In the analysis of the tutor’s discourse subject, the discourse subject in the sample with the lowest amount of tutor’s discourse still reaches 58.845%, and the average lexical density of tutors in the selected 40 discourse samples is 0.3732. In general, tutors’ discourse is higher, and the discourse weights of counselors and students should be evenly distributed when constructing the counselor discourse system.</p>
</abstract>

As an important guide in the growth process of college students, how to communicate effectively with college students in discourse is a problem faced by college counselors. The LDA college counselor discourse text analysis model is created using Word2vec after processing the discourse of college counselors through text conversion in this paper. The model improves the topic resolution of the LDA model by removing the deactivated words in the discourse text and enriches the word vector data of the discourse text by combining it with the function of expanding feature words of Word2vec. The constructed model is applied to analyze the current situation of the tutor’s discourse themes and subjects by combining the topics generated during the discourse exchange between the tutor and the student and the corresponding generation probability and constructing the tutor discourse system in colleges and universities according to the analysis results. In the analysis of discourse topics, the generation probability of the topic word of personal emotion is the largest among the 20 topics, which is 0.502498, exceeding 0.5. In the analysis of the tutor’s discourse subject, the discourse subject in the sample with the lowest amount of tutor’s discourse still reaches 58.845%, and the average lexical density of tutors in the selected 40 discourse samples is 0.3732. In general, tutors’ discourse is higher, and the discourse weights of counselors and students should be evenly distributed when constructing the counselor discourse system.

The Construction and Transformation of the Discourse System of College Counselors in the New Media Era

Applied Mathematics and Nonlinear Sciences

This work is licensed under the Creative Commons Attribution 4.0 International License.

As an important guide in the growth process of college students, how to communicate effectively with college students in discourse is a problem faced by...

The Construction and Transformation of the Discourse System of College Counselors in the New Media Era

Publié en ligne: 31 janv. 2024

Pages: -

Reçu: 18 déc. 2023

Accepté: 24 déc. 2023

DOI: https://doi.org/10.2478/amns-2024-0159

Mots clés
Word2vec, LDA model, Topic resolution, Expanded feature words, Discourse system construction

© 2024 Li Liu, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

The Construction and Transformation of the Discourse System of College Counselors in the New Media Era

Publié en ligne: 31 janv. 2024

Pages: -

Reçu: 18 déc. 2023

Accepté: 24 déc. 2023

DOI: https://doi.org/10.2478/amns-2024-0159

Mots clésWord2vec, LDA model, Topic resolution, Expanded feature words, Discourse system construction

© 2024 Li Liu, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Mots clés
Word2vec, LDA model, Topic resolution, Expanded feature words, Discourse system construction