Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/50909
Title: Estimation of rumor sources in social network
Authors: Purnama, Evan.
Keywords: DRNTU::Engineering::Electrical and electronic engineering::Computer hardware, software and systems
Issue Date: 2012
Abstract: Social network was becoming very popular in these couple years. More people got into social network and more people had capability to spread and pass the information in the Internet. It became more difficult for people to indicate whether the information over the Internet was reliable. The sources of the information were commonly difficult to be determined. The objective of this project is to study a social network analysis and study the method to find the influential source of information in the social network by using real data from one of the biggest and most impactful social networking sites. More than 1 million data from Twitter’s firehouse dataset were taken and being analyzed based on the relationship on each user. The relationship of users that being analyzed were Retweet, Mention/Reply, Indirect Retweet, Indirect Mention. The different weighting systems were implemented in each type of relationship. The results of the project showed that users from the predicted user lists appear in the testing results. The top 1,000 out of 761,631 predicted users (which were the sources of the topic) were taken and being compared. It showed more than 250 users appear in the test results. The experiment and project also showed that Twitter’s Streaming API was suitable for fetching large and extensive data.
URI: http://hdl.handle.net/10356/50909
Rights: Nanyang Technological University
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:EEE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
eA5152.pdf
  Restricted Access
Main FYP Report2.55 MBAdobe PDFView/Open

Google ScholarTM

Check

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.