Work

AIGC research on Content innovation and Enhancing User engagement in social media marketing

Year: 2025

Type: conference-paper

Abstract: Based on the theory of information ecology, this study builds a model of influencing factors of user interaction behavior in AIGC (Generative Artificial Intelligence) application platform. The model was verified by questionnaire survey data to analyze the significant positive effects of immersion experience, self-efficacy and community influence on users' persistent use behavior. Using K-means clustering algorithm to generate user portraits, combined with generative adversarial network (GAN) and bidirectional encoder representation (BERT) model, especially GAN-BERT model, intelligent conversation function of virtual anchor is realized. The test results show that GAN-BERT model performs well in natural language processing tasks, with a BLEU value of 44.25, which significantly improves the naturalness and accuracy of dialogue. In addition, AIGC technology effectively reduces the creation cost of merchants and improves the creation efficiency by automatically generating high-quality text, pictures and video content.

Source: International Conference on Computer Information and Big Data Applications (CIBDA)

DOI: 10.1145/3746709.3746904

Authors: Tao Zhou, Xinxing Luo, Xinyu Yang, Sheng Cao

Language: English

Topic: Digital Marketing and Social Media

Subfield: Sociology and Political Science

Field: Social Sciences

Domain: Social Sciences

Open Access status: hybrid

Funders: Hubei Provincial Department of Education, Wuhan Donghu University

OpenAlex ID: W4413812466

FWCI: 3.432

Cites: 8

Cited by: 1

Related to: 10