Analysis of twitter to identify trends and influentials with a case study on Turkish twitter users

Göktürk, Gökhan (2014) Analysis of twitter to identify trends and influentials with a case study on Turkish twitter users. [Thesis]

[thumbnail of GokhanGokturk_10049671.pdf] PDF
GokhanGokturk_10049671.pdf

Download (2MB)

Abstract

Social media is one of the largest information flow medium today. Nevertheless, despite its centrality, conventional public opinion research doesn't take social media into account but instead focuses on surveys, polls and interviews. These research methods have their limitations. By nature, even the most meticulously designed survey, for example, is limited by time and seldom bias free. If properly utilized social media, can address limitations of these shortcomings; Social Media allows us to continuously observe how information flows both temporally and spatially since its users communicate with each other rather than answering survey questions; the data is without experimenter bias and sample size is much larger than of conventional methods. We aimed to show an interdisciplinary work that provides empirical quantifiable answers for social science problems using network analysis and machine learning. With this aim in mind, this work combines network analysis and sentiment analysis to analyze Istanbul 2014 local elections as a proof of concept. Furthermore, it illustrates the performance of our sentiment analysis system and structural differences between two parties in the event.
Item Type: Thesis
Additional Information: Yükseköğretim Kurulu Tez Merkezi Tez No: 394269.
Uncontrolled Keywords: Social Network Analysis. -- Sentiment Analysis. -- Text Classification.
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7800-8360 Electronics > TK7885-7895 Computer engineering. Computer hardware
Divisions: Faculty of Engineering and Natural Sciences > Academic programs > Computer Science & Eng.
Faculty of Engineering and Natural Sciences
Depositing User: IC-Cataloging
Date Deposited: 25 Apr 2018 11:20
Last Modified: 26 Apr 2022 10:17
URI: https://research.sabanciuniv.edu/id/eprint/34523

Actions (login required)

View Item
View Item