Estimation of rumor sources in social network
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 o...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project |
| Language: | English |
| Published: |
2012
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/50909 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | 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. |
|---|