Using Word Embeddings to Analyze Protests News
This work addresses incremental improvements in text classification for protest news analysis, benefiting researchers in computational social science.
The study tackled the problem of classifying protest-related news by replacing existing word embeddings with ELMo and DistilBERT in CLEF 2019 tasks, resulting in DistilBERT improving the F1-score to 0.66 compared to FastText and outperforming ELMo across both tasks.
The first two tasks of the CLEF 2019 ProtestNews events focused on distinguishing between protest and non-protest related news articles and sentences in a binary classification task. Among the submissions, two well performing models have been chosen in order to replace the existing word embeddings word2vec and FastTest with ELMo and DistilBERT. Unlike bag of words or earlier vector approaches, ELMo and DistilBERT represent words as a sequence of vectors by capturing the meaning based on contextual information in the text. Without changing the architecture of the original models other than the word embeddings, the implementation of DistilBERT improved the performance measured on the F1-Score of 0.66 compared to the FastText implementation. DistilBERT also outperformed ELMo in both tasks and models. Cleaning the datasets by removing stopwords and lemmatizing the words has been shown to make the models more generalizable across different contexts when training on a dataset with Indian news articles and evaluating the models on a dataset with news articles from China.