Implementation of Query Expansion to Enhance Word2Vec Performance in Hoax News Detection Systems

The rapid development of the internet has made accessing various news easy and quick. However, the spread of news on the internet sometimes does not align with the actual situation, leading to the need for hoax news detection systems. These systems are built to analyze the truthfulness of news articles. One popular approach in hoax news detection is the utilization of the Word2Vec model for feature extraction. Unfortunately, the Word2Vec model has a limitation in representing words that are not present in the training data, also known as out-of-vocabulary (OOV) words. To address this weakness, this research proposes the application of the XLN et transfer learning model combined with the Query Expansion technique. By incorporating XLNet, the research aims to solve the OOV problem by expanding the original words while preserving the overall context of the words. This approach is expected to improve the performance of Word2Vec in the hoax news detection system. In summary, this research focuses on enhancing the hoax news detection system by leveraging the XLNet transfer learning model and Query Expansion technique to tackle the OOV problem inherent in the Word2Vec model. By doing so, the aim is to improve the accuracy and effectiveness of the system in identifying and classifying hoax news articles.