Sentiment Analysis of Turkish and English Twitter Feeds Using Word2Vec Model
Abdullah Ammar Karcıoğlu, Tolga Aydın · 2019
Social media has become an important part of daily life. With twitter, one of the most popular social media services, users express their feelings and thoughts to the whole world using twitter posts. For this reason, twitter feeds have become an important source of sentiment analysis. In this study, the apply of Word2Vec model in the classification of labeled data in English and Turkish Twitter feeds and the effect of getting root on feeds to Word2Vec model are investigated. Our study has two different data sets, English and Turkish. BOW and Word2Vec models were applied to each data set in the case where twitter feeds were not get roots and get roots were extracted. In this study, which is implemented in the Python programming language, the success percentages are compared by applying the scikit-learn classification algorithms, Linear SVM and Logistic Regression.