TRCM: A Methodology for Temporal Analysis of Evolving Concepts in Twitter

Abstract

The Twitter network has been labelled the most commonly used microblogging application around today. With about 500 million estimated registered users as of June, 2012, Twitter has become a credible medium of sentiment/opinion expression. It is also a notable medium for information dissemination; including breaking news on diverse issues since it was launched in 2007. Many organisations, individuals and even government bodies follow activities on the network in order to obtain knowledge on how their audience reacts to tweets that affect them. We can use postings on Twitter (known as tweets) to analyse patterns associated with events by detecting the dynamics of the tweets. A common way of labelling a tweet is by including a number of hashtags that describe its contents. Association Rule Mining can find the likelihood of co-occurrence of hashtags. In this paper, we propose the use of temporal Association Rule Mining to detect rule dynamics, and consequently dynamics of tweets. We coined our methodology Transaction-based Rule Change Mining (TRCM). A number of patterns are identifiable in these rule dynamics including, new rules, emerging rules, unexpected rules and 'dead' rules. Also the linkage between the different types of rule dynamics is investigated experimentally in this paper.

An Overview on the Use of Data Mining and Linguistics Techniques for Building Microblog-Based Early Detection Systems in the Healthcare Sector

“…OSNs, such as Twitter is just one, but an important platform that disseminates news and events in nearly real-time [8]. Twitter has become a very important medium of opinion expression and information propagation on varied topics [9]. Twitter users can write tweets and follow other users' tweets from anywhere in the world by using mobile devices or simply desktop computers.…”

Extraction of Unexpected Rules from Twitter Hashtags and its Application to Sport Events

“…Traditional newsagents follow the activities on Twitter to assist them in updating related news [15]. Many Topic Detection and Tracking TDT researches have been conducted on Twitter data [1], [2], [6], [7], [13], [14], [16]- [18], [22], [25], the results of which show the dynamism of Twitter network as information dissemination tool and the efficiency of data mining techniques in extracting newsworthy contents from tweets posted online. Twitter is known to have played major role in publicising and sustaining the trend of major news of global relevance on the network.…”

A Survey of Data Mining Techniques for Internet

“…The outcome of the experiments shows that peak detection using Continuous Wavelet Transformation realized impressive outcomes in the ascertaining abrupt increases on the mention of specific hashtags. In [1] the abruptness in hashtags usage is labelled unexpected rule evolvement which is discussed under TRCM in Section 8.1.…”

“…Using the experimental study result in [37], it was substantiated that the lifespan of tweets" hashtags can be related to evolvements of news and events in reality. The TRCM techniques [1] and [37] can be said to be the first time ARM was used to mine Twitter data. This therefore opens up the area for further research as ARM can be fine-tuned to be used to mine data on other social networks sites for information retrieval and knowledge acquisition.…”

“…Association Rule Mining (ARM) can find the probability of co-existence of tweets" hashtags. Firstly, in [1] ARM was used to analyse tweets on the same topic over consecutive time periods t and t+1. Rule Matching (RM) was later employed to detected changes in patterns such as "emerging (EM)", "unexpected consequent (UnxCs)" and "unexpected conditional (UnxCn)", "new (N)" and "de ad (D)" rules in tweets.…”