Social network analytics : trend analysis in Twitter

• Main Author: Soh, Jophia Yi Wen
• Other Authors: Sourav Saha Bhowmick
• Format: Final Year Project
• Language: English
• Published: 2017
• Subjects: DRNTU::Engineering::Computer science and engineering::Software::Software engineering
• Online Access: http://hdl.handle.net/10356/70453
• Institution: Nanyang Technological University

Twitter is one of a popular microblogging site that allows registered user to engage and interact in short messages pertaining to wide variety of topics, not exceeding 140 characters. Alongside other social networking services, Twitter adopt a social model called “following” which allows registered users to follow and receive updates from the registered account they follow, in real-time. As a result, this social model has enabled Twitter to become very efficient in information spreading and propagation. However, with many news, sources and topics discussed on Twitter, it brings about huge drawbacks on the efficient acquisition of important information. It is very time consuming to look through tremendous set of tweets to obtain important information. In this proposed project, trend analysis is performed to acquire latest trends in the most recent collection of tweets to overcome this limitation as a key to facilitate the search of significant information among huge number of tweets. With an existing topic modelling-based recent tweet summarizer, we built beyond and perform trend analysis on the tweets across time and topic to provide the latest emerging trends summarised in topics, to provide quick overview to the users. A collection of 800 most recent English tweets localised with user mentioned keywords were considered and implemented using the most fundamental approach known as hashtag implementation. To further handle the popular hashtags in real-time, we extend the implementation by incorporating Linkify to the group of weighted hashtags.