@inproceedings{f1167b69b7ea4239bc1cb1a0a402fb2b,
title = "Eigenspace-based fuzzy c-means for sensing trending topics in Twitter",
abstract = "As the information and communication technology are developed, the fulfillment of information can be obtained through social media, like Twitter. The enormous number of internet users has triggered fast and large data flow, thus making the manual analysis is difficult or even impossible. An automated methods for data analysis is needed, one of which is the topic detection and tracking. An alternative method other than latent Dirichlet allocation (LDA) is a soft clustering approach using Fuzzy C-Means (FCM). FCM meets the assumption that a document may consist of several topics. However, FCM works well in low-dimensional data but fails in high-dimensional data. Therefore, we propose an approach where FCM works on low-dimensional data by reducing the data using singular value decomposition (SVD). Our simulations show that this approach gives better accuracies in term of topic recall than LDA for sensing trending topic in Twitter about an event.",
keywords = "clustering, fuzzy c-means, singular value decomposition, topic detection, topic modeling",
author = "T. Muliawati and Hendri Murfi",
note = "Publisher Copyright: {\textcopyright} 2017 Author(s).; 2nd International Symposium on Current Progress in Mathematics and Sciences 2016, ISCPMS 2016 ; Conference date: 01-11-2016 Through 02-11-2016",
year = "2017",
month = jul,
day = "10",
doi = "10.1063/1.4991244",
language = "English",
series = "AIP Conference Proceedings",
publisher = "American Institute of Physics Inc.",
editor = "Sugeng, {Kiki Ariyanti} and Djoko Triyono and Terry Mart",
booktitle = "International Symposium on Current Progress in Mathematics and Sciences 2016, ISCPMS 2016",
}