Enhancing Opinion Mining in Social Networks using Graph-based Analysis

Please enable javascript in your browser.

Hosseini, Mohammad Hamed | 2018

851 Viewed

Type of Document: M.Sc. Thesis
Language: Farsi
Document No: 51400 (02)
University: Sharif University of Technology
Department: Mathematical Sciences
Advisor(s): Moghadasi, Reza
Abstract:
In the past, opinion polls have generally been conducted using statistical techniques and the comments of those samples, which, in addition to its various challenges, require much power and cost per poll. Today, due to the high penetration rate of social networks in society, a very large fraction of people discuss such issues in terms of their opinions and preferences in these networks, and hence, these networks can be a valuable source of information to get the opinions of people in a short time, and by lower cost.The purpose of this thesis is to enhancing the results of the surveys which have already performed using various methods of text processing in social networks by using the graph structure of these networks in a way that opinion mining(i.e., labeling of users in positive, negative, neutral) on a specific issue, get better results by analysing the connections of each user with other users and with other social network entities
Keywords:
Social Networks ; Sentiment Analysis ; Opinion Mining ; Graph Analysis ; Relational Learning

مقدمات
- تعاریف اولیه و نمادگذاری
  - ماتریس مجاورت
  - ماتریس لاپلاسین
  - ماتریس انتقال
- صورت‌بندی مسئله
- رویکردهای مختلف حل مسئله
  - خوشه‌بندی جمعی
  - یادگیری شبه‌نظارتی مبتنی بر گراف
  - مقایسه دو رویکرد
ویژگی‌های رئوس
- روش Moran’s I
- روش Geary’s c
- روش نسبت مجاورت کمینه
- روش تجزیه ماتریس مجاورت
قرابت رئوس
- قرابت ویژگی‌های رئوس
- شاخص‌های قرابت ساختاری
  - شاخص‌های موضعی
  - شاخص‌های سراسری
طراحی مدل یادگیری
- طراحی ویژگی‌ها
  - استخراج ویژگی از روی قرابت رئوس
  - طراحی ویژگی‌های نهایی
- انتخاب مدل یادگیری
  - Logistic Regression
  - Naïve Bayes
  - Support Vector Machine
جمع‌بندی و مسائل پیش‌رو