Sentiment Analysis of Twitter using Word Embeddings in Python

Today’s Internet content is largely made from unstructured data, mostly images and text. Part of the text content is in form of comments made by users giving their impressions about virtually everything. Processing natural language text documents is still a challenging task due to its ambiguity and context dependency. Nevertheless, being able to extract useful information from this source is highly desirable due to its market value. Understand better people’s feeling about a matter, customer needs, or product’s quality, are just some of the possibilities by accomplishing this task. In this work, we evaluate the performance of machine learning and natural language pro- cessing techniques in the task of classifying sentiment in tweets.