Loading...

Persian Aspect-based Sentiment Analysis using Unsupervised Learning Methods

Akhondzadeh, Reza | 2017

579 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 49493 (19)
  4. University: Sharif University of Technology
  5. Department: Computer Engineering
  6. Advisor(s): Ghasem-Sani, Gholamreza
  7. Abstract:
  8. Sentiment analysis, is a subfield of natural language processing that aims at opinion mining to analyze thoughts, orientation and evaluation of users within some texts. Different organizations in multiple social domains, use this approach as a tool to asses their strengths and shortcomings. In sentiment analysis, the goal is to use machine learning techniques with the purpose of specifying users’ positive or negative orientation about a product or merchandise. The solution to this problem includes two main steps: extracting aspects and determining users’ positive or negative sentiments in respect to the aspects. Two main challenges of sentiment analysis in Farsi, are lack of comprehensive tagged data sets and use of colloquial language in texts. One way to overcome the first challenge is to use unsupervised methods. In this dissertation, at first, different methods of dealing with sentiment analysis problem are examined. Next an aspect based sentiment analysis system in Farsi, using unsupervised methods, is introduced. The aspect extraction step is done using topic modeling methods and neural networks with the help of rule based methods. Word sentiment extraction and positive or negative orientation in respect to aspects is accomplished by rule-based methods and distance metrics. This system is evaluated by using precision and recall metrics. F1 metric for aspect words and sentiment words in the proposed system are 0/64 and 0/56, respectively
  9. Keywords:
  10. Sentiment Analysis ; Unsupervised Method ; Topic Modeling ; Neural Network ; Accuracy ; Rule Based System ; Aspect Mining

 Digital Object List

 Bookmark

No TOC