HAHA at FakeDeS 2021: A Fake News Detection Method Based on TF-IDF and Ensemble Machine Learning

Kun Li
2021 Annual Conference of the Spanish Society for Natural Language Processing  
This paper describes our participation in the FakeDeS [5] Task at Iberlef 2021: Fake News Detection in Spanish. Base on this task, we propose the classic TF-IDF feature extraction technology and Stacking ensemble learning method base on weak classifiers. It not only analyzes the content of the news, but also combines effective information such as publishers and topics to improve the performance of our model. We used five machine learning models, and achieved very competitive results on both the
more » ... validation set and the test set, and got the second place in the final evaluation phase.
dblp:conf/sepln/Li21a fatcat:54nkvi2355axrdk4wnciabc56a