Impact of Fine-Tuned Embeddings on Machine Learning-Based Causal Detection

Causal detection from text is a crucial area of research in the field of natural language processing. In this research, traditional word embedding techniques are integrated with multiple classifiers for causal detection. Fine-tuning of traditional word embedding techniques was implemented to enhance prediction efficiency. This study evaluates the performance of TF-IDF-XGBoost, BoW-XGBoost, Word2Vec-SVM Polynomial, FastText-SVM Polynomial, Skipgram-SVM Polynomial, GloveSVM Linear. This research highlights the importance of choosing the appropriate classifier and embedding techniques in the text classification task. The TF-IDF with XGBoost was found to be the most appropriate combination with accuracy, and the recorded F1 score is $97.2136 \%$ and $97.216 \%$, respectively.