Authorship Verification (AV) is one of the interesting topics that had developed rapidly and distinctly since the middle of the 19th century. With the social media era, there is always a problem in determining whether a given tweet, post, or comment was written by a certain user or not. We are proposing a new approach to verify if a tweet belongs to a claimed user. Our proposed method utilizes the benefits of one-shot learning. It is based on vectors similarity which depends on Term Frequency?Inverse Document Frequency (TF-IDF) and word embedding for better verification accuracy. After comparisons, our proposed approach outperforms existing methods in the case of cross topics.
展开▼
