Opinion Mining and Sentiment Analysis using Bayesian and Neural Networks Approaches
Name
Olha Shepelenko
Abstract
Information technologies have firmly entered our life and it is impossible to imagine our life without gadgets or the Internet. Today, social media is not only a source that broadcasts information to the users, but it allows users to intercommunicate and share their views and experience with each other. Some portion of such data is subjective and contains opinionated information that can be further analyzed to retrieve essential data from it and later use for various purposes for analysis and decision support. In order to use this type of that the first step is to understand it and categorize opinions in the information. Hence, in this dissertation, sentiment analysis techniques are studied in order to retrieve opinions from the tweets. In order to ensure efficient classification, it is important to apply algorithms that perform well on this task. Therefore, the main goal of the thesis is to investigate algorithms that can be applied for the opinion estimation. To that extend, data preprocessing and several experiments are conducted, namely, the classifier is trained and tested on two different datasets with two different classifiers (Naive Bayes and convolutional neural network). In addition, the influence of the training data on the classifier efficiency is discussed.
Graduation Thesis language
English
Graduation Thesis type
Master - Computer Science
Supervisor(s)
Amnir Hadachi, Artjom Lind
Defence year
2017