EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Feature fusion based transformer for sentiment analysis in social networks

Shiyong Li, He Li iD, Juan Du, Shitao Yan, Chuang Dong

PLOS One · 07.11.2025 · 20 (11) · e0333416

DOI: 10.1371/journal.pone.0333416 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, online erschienen am 07.11.2025. Verfasst von Shiyong Li und 4 weiteren Autor:innen. Beteiligte Einrichtung: School of Computer Science and Technology, Henan Institute of Science and Technology, Xinxiang, China. Hauptthemen laut MEDLINE: Social Media, Social Networking. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 50 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Key Scientific and Technological Research Projects in Henan Province, National Natural Science Foundation of China, Training Plan for Young Backbone Teachers in Higher Education Institutions in Henan Province.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Sentiment analysis methods aim to evaluate users’ mental health conditions by analyzing their posted content (text, images, and audio) on social networks. However, given the diversity and complexity of social media information, traditional single-modal sentiment analysis techniques exhibit limitations in accurately interpreting users’ emotional states and may even lead to contradictory conclusions. To address this challenge, this paper proposes a Feature Fusion Based Transformer (FFBT) solution. The framework consists of three key steps: Firstly, RoBERTa and ResNet50 models are employed to extract features from textual and image data in social media posts, respectively. Then, a multimodal Transformer architecture facilitates feature alignment and fusion across different modalities. Finally, the fused features are fed into a fully connected network (FCN) for sentiment classification, ultimately determining the user’s emotional state. Experiments conducted on a custom dataset constructed from social media platform data demonstrate that FFBT outperforms existing sentiment analysis algorithms by 4.1% in accuracy and 5% in F1-scores, respectively.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Social MediaSocial NetworkingHumansAlgorithmsEmotions

Autor:innen und Institutionen

  • School of Computer Science and Technology, Henan Institute of Science and Technology, Xinxiang, China.
  • Henan Intelligent Emergency Service and Security Engineering Research Center, Nanyang Normal University, Nanyang, China.
  • Business School, Xinxiang University, Xinxiang, China.

ORCID-Profile: He Li

Förderung

  • Key Scientific and Technological Research Projects in Henan Province
  • National Natural Science Foundation of China
  • Training Plan for Young Backbone Teachers in Higher Education Institutions in Henan Province
  • Key Scientific Research Projects of Colleges and Universities in Henan Province

Zitieren

APA

Li, S., Li, H., Du, J., Yan, S., & Dong, C. (2025). Feature fusion based transformer for sentiment analysis in social networks. PLOS One, 20(11), e0333416. https://doi.org/10.1371/journal.pone.0333416

Vancouver

Li S, Li H, Du J, Yan S, Dong C. Feature fusion based transformer for sentiment analysis in social networks. PLOS One. 2025;20(11):e0333416. doi:10.1371/journal.pone.0333416

BibTeX
@article{Li2025,
  author = {Li, Shiyong and Li, He and Du, Juan and Yan, Shitao and Dong, Chuang},
  title = {Feature fusion based transformer for sentiment analysis in social networks},
  journal = {PLOS One},
  year = {2025},
  volume = {20},
  number = {11},
  pages = {e0333416},
  doi = {10.1371/journal.pone.0333416}
}
Kostenlos starten →