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

<p>With the rapid development of information media, senior counselors face many new opportunities and challenges in ideological and political education discourse. This study divides the counselor discourse modal system into linguistic and visual image modal. Adopting the advanced Transformer architecture and combining the BERT and RoBERTa algorithms, this paper profoundly analyzes the syntactic structure in the discourse. It extracts the emotional features in the text using the bert word vector model. Meanwhile, the dynamic features in the audio information of the discourse are extracted by MFCC technique. After completing the preprocessing of discourse features, we used Bayesian classifier for classification and recognition to further refine the annotation of discourse information. The study results showed that the overall sentiment of the discourse of tutor number 1 tended to be harmful during the 30 minutes, with six segments having an emotional pleasantness of more than -2. In addition, the density of positive discourse sentiment of tutor number 3 was 0.0811 higher than that of tutor number 4. In contrast, the number of discourse labeling of tutor number 5 was reduced by 74 times. Through in-depth analysis and understanding of these discourse modalities and affective features, tutors can better adapt to the information media environment and effectively carry out ideological and political education work.</p>
</abstract>

With the rapid development of information media, senior counselors face many new opportunities and challenges in ideological and political education discourse. This study divides the counselor discourse modal system into linguistic and visual image modal. Adopting the advanced Transformer architecture and combining the BERT and RoBERTa algorithms, this paper profoundly analyzes the syntactic structure in the discourse. It extracts the emotional features in the text using the bert word vector model. Meanwhile, the dynamic features in the audio information of the discourse are extracted by MFCC technique. After completing the preprocessing of discourse features, we used Bayesian classifier for classification and recognition to further refine the annotation of discourse information. The study results showed that the overall sentiment of the discourse of tutor number 1 tended to be harmful during the 30 minutes, with six segments having an emotional pleasantness of more than -2. In addition, the density of positive discourse sentiment of tutor number 3 was 0.0811 higher than that of tutor number 4. In contrast, the number of discourse labeling of tutor number 5 was reduced by 74 times. Through in-depth analysis and understanding of these discourse modalities and affective features, tutors can better adapt to the information media environment and effectively carry out ideological and political education work.

The Construction and Transformation of the Discourse System of Higher Vocational Education Counselors in the Era of Information Education

Applied Mathematics and Nonlinear Sciences

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

With the rapid development of information media, senior counselors face many new opportunities and challenges in ideological and political education discourse....

The Construction and Transformation of the Discourse System of Higher Vocational Education Counselors in the Era of Information Education

Publicado en línea: 26 feb 2024

Páginas: -

Recibido: 10 ene 2024

Aceptado: 18 ene 2024

DOI: https://doi.org/10.2478/amns-2024-0466

Palabras clave
Transformer, Bert word vector, MFCC discourse feature extraction, Discourse emotional state, Discourse system construction

© 2024 Nana Gao et al., 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 Higher Vocational Education Counselors in the Era of Information Education

Publicado en línea: 26 feb 2024

Páginas: -

Recibido: 10 ene 2024

Aceptado: 18 ene 2024

DOI: https://doi.org/10.2478/amns-2024-0466

Palabras claveTransformer, Bert word vector, MFCC discourse feature extraction, Discourse emotional state, Discourse system construction

© 2024 Nana Gao et al., published by Sciendo

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

Palabras clave
Transformer, Bert word vector, MFCC discourse feature extraction, Discourse emotional state, Discourse system construction