EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Fine-Grained Sentiment Quantification of Media Texts Considering Sentence Type and Holder–Target Awareness

Xiaoqing Ju, Daichao Li, Yunqiang Zhu, Li Yu, Qiang Wang, Yichen Yang, Renwen Xia, Peng Ye, Shu Wang

ISPRS International Journal of Geo-Information · 07.05.2026 · 15 (5) · 201

DOI: 10.3390/ijgi15050201 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ISPRS International Journal of Geo-Information, online erschienen am 07.05.2026. Verfasst von Xiaoqing Ju und 8 weiteren Autor:innen. Die Arbeit nennt 25 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, National Key R&D Program of China, Chinese Academy of Sciences.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Media texts convey emotions and stances that can shape the evolution of public opinion, which calls for comparable quantitative sentiment analysis. However, most existing approaches assign a single sentiment score to an entire article, making it difficult to distinguish functional differences across sentences and to clarify who expresses the sentiment and what the evaluation targets, thereby limiting interpretability and cross-source comparability. To address this issue, we propose a fine-grained sentiment quantification method for media texts that jointly considers sentence types and opinion holder–target structure. The method obtains sentence-level sentiment scores and simultaneously extracts sentence types, opinion holders, and opinion targets, enabling article-level structured quantification and comparison under a unified evaluation setting. In our implementation, a large language model (LLM) is primarily used for semantic parsing and structured extraction. Experiments demonstrate that the proposed method delivers stable performance on the sentiment score regression task (R2 = 0.899, MAE = 0.088, MSE = 0.027; relative to the strongest fine-tuned pretrained language model baseline in our comparison, RoBERTa with R2 = 0.871, this corresponds to a 2.8-percentage-point gain in R2 and an 8.3% reduction in MAE), and effectively supports opinion holder–target identification (holder weighted average F1 = 0.812; target loose F1 = 0.691 in a supplementary evaluation). Building on these outputs, the method can further reveal the spatial distribution of sentiment bias in global media coverage, highlighting relative sentiment patterns in cross-national narratives.

Förderung

  • National Natural Science Foundation of China
  • National Key R&D Program of China
  • Chinese Academy of Sciences

Zitieren

APA

Ju, X., Li, D., Zhu, Y., Yu, L., Wang, Q., Yang, Y., Xia, R., Ye, P., & Wang, S. (2026). Fine-Grained Sentiment Quantification of Media Texts Considering Sentence Type and Holder–Target Awareness. ISPRS International Journal of Geo-Information, 15(5), 201. https://doi.org/10.3390/ijgi15050201

Vancouver

Ju X, Li D, Zhu Y, Yu L, Wang Q, Yang Y, et al. Fine-Grained Sentiment Quantification of Media Texts Considering Sentence Type and Holder–Target Awareness. ISPRS International Journal of Geo-Information. 2026;15(5):201. doi:10.3390/ijgi15050201

BibTeX
@article{Ju2026,
  author = {Ju, Xiaoqing and Li, Daichao and Zhu, Yunqiang and Yu, Li and Wang, Qiang and Yang, Yichen and Xia, Renwen and Ye, Peng and Wang, Shu},
  title = {Fine-Grained Sentiment Quantification of Media Texts Considering Sentence Type and Holder–Target Awareness},
  journal = {ISPRS International Journal of Geo-Information},
  year = {2026},
  volume = {15},
  number = {5},
  pages = {201},
  doi = {10.3390/ijgi15050201}
}
Kostenlos starten →