Survey on clustering techniques in Twitter data
Authors:
- 1 February 2018
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1073-1077
Abstract
Social networks and online news services are used by users to communicate and share messages. One such social network is Twitter. Earlier its messages were restricted to 140 characters, but from November 7, 2017 its limit was extended to 280 characters except Japanese, Korean and Chinese languages. Because of restricted characters used, it is famously called micro blogging. Mining twitter data has become popular, because it provides useful information which is being used in various fields. This paper highlights various clustering techniques that can be used in twitter data mining with advantages and limitation.Keywords
This publication has 6 references indexed in Scilit:
- A Survey of Partition based Clustering Algorithms in Data Mining: An Experimental ApproachInformation Technology Journal, 2011
- A simple and fast algorithm for K-medoids clusteringExpert Systems with Applications, 2009
- Survey of Clustering AlgorithmsIEEE Transactions on Neural Networks, 2005
- Why so many clustering algorithmsACM SIGKDD Explorations Newsletter, 2002
- A survey of fuzzy clusteringMathematical and Computer Modelling, 1993
- A Survey of Recent Advances in Hierarchical Clustering AlgorithmsThe Computer Journal, 1983