Bilkent University
Department of Computer Engineering


Aspect Based Opinion Mining on Turkish Tweets


Esra Akbaş
MSc Student
Computer Engineering Department
Bilkent University

Understanding opinions about entities or brands is instrumental in reputation management and decision making. With the advent of social media, more people are willing to publicly share their recommendations and opinions. As the type and amount of such venues increase, automated analysis of sentiment on textual resources has become an essential data mining task. Sentiment classification aims to identify the polarity of sentiment in text. The polarity is predicted on either a binary (positive, negative) or a multi-variant scale as the strength of sentiment expressed. Text often contains a mix of positive and negative sentiments, hence it is often necessary to detect both simultaneously.

In this thesis, we investigate the problem of mining opinions by extracting aspects of entities/topics on collection of short texts. We focus on Turkish tweets that contain informal short messages. Most of the available resources such as lexicons and labeled corpus in the literature are for the English language. Our approach would help enhance the sentiment analyses to other languages where such rich sources do not exist. After a set of preprocessing steps and automated generation of a list of Turkish words with their sentiment strengths, we design feature vectors to detect mixture of positive and negative sentiments. We adapt machine learning methods to generate classifiers based on these multi-variant scale features, and test their performance on a Turkish tweet data collected over time via Twitter API.


DATE: 6 July, 2012, Friday @ 14:00