Fraunhofer SIT at CheckThat! 2022: Semi-Supervised Ensemble Classification for Detecting Check-Worthy Tweets
2022 Conference and Labs of the Evaluation Forum, CLEF 2022
; 3180:500-510, 2022.
Article
in English
| Scopus | ID: covidwho-2011389
ABSTRACT
During the corona pandemic misinformation has been increasingly spread on social media. Since the automatic verification of social media postings has shown to be challenging, there exists the need of classification systems, that can identify check-worthy posts in social media feeds. In this paper, the classification system used in the CLEF2022-CheckThat! Lab to detect check-worthiness in English tweets is presented. The proposed system, that took advantage of ensemble and semi-supervised learning showed promising results in the experimental evaluation. Further, first experiments were conducted with the novel framework lambeq to solve the classification problem using quantum natural language processing (QNLP), which were not part of the final model. The final model ranked fifth best in terms of the F1-score in the competition. © 2022 Copyright for this paper by its authors.
check-worthiness detection; GAN; QNLP; semi-supervised learning; Twitter; Machine learning; Natural language processing systems; Social networking (online); Classification system; Fraunhofer; Language processing; Natural languages; Quantum natural language processing; Social media; Learning algorithms
Search on Google
Collection:
Databases of international organizations
Database:
Scopus
Language:
English
Journal:
2022 Conference and Labs of the Evaluation Forum, CLEF 2022
Year:
2022
Document Type:
Article
Similar
MEDLINE
...
LILACS
LIS