Uncorrected Author Proof Journal of Intelligent & Fuzzy Systems xx (20xx) x–xx DOI:10.3233/JIFS-169959 IOS Press 1 Information entropy based event detection during disaster in cyber-social networks 1 2 A. Bhuvaneswari ∗ and C. Valliyammai 3 Department of Computer Technology, Madras Institute of Technology, Anna University, Chennai 4 Abstract. The demand for Cyber Social Networks has increasingly become the main source of information propagation due to the rapid growth of micro-blogging activity between socially connected people. The process of detecting disaster events, in huge volumes, on fast-streaming platform is quite challenging. In this paper, an information entropy based event detection framework is proposed to identify the event and its location by clustering relatively high-density ratio of tweets using Twitter data. The Shannon entropy of target users, location, time intervals and hashtags are estimated to quantify the dissemination of events as “how-far about” in real- world using entropy maximization inference model. The geo-tagged (spatial) tweets are extracted for a specified time period (temporal) to identify the location of an event as “where-when about”; and visualizes the event in geo-maps. The evaluation parameters of Entropy, Cluster Score, Event Detection Hit and False Panic Rate during four major disaster events are identified to illustrate the effectiveness of the proposed framework. The retweeting activity of the Twitter user is classified as human signatures and bots. The experimental outcome determines the scope and significant dissemination direction of finding events from a new perspective which demonstrates 96% of improved event detection accuracy. 5 6 7 8 9 10 11 12 13 14 15 16 Keywords: Cyber-social networks, event detection, geo-tag, spatiotemporal, Shannon entropy 17 1. Introduction 18 In recent years, Cyber-Social Networking (CSN) 19 platforms like Twitter, Facebook, Sina Weibo and 20 Tumblr have encapsulated huge chunks of infor- 21 mation comprising insightful user opinions about 22 various events [1]. The problem of evaluating 23 unstructured data in user generated content has ubiq- 24 uitous applications, including the identification of 25 conversation topics and abnormal events [2]. The 26 online events that match with actual real-time events 27 are of diverse temporal and spatial scales, particularly 28 disasters like earthquakes, floods, landslides, etc. to 29 deliver prior alerts, support, and immediate recovery 30 [3, 4]. The CSN act as communication tool and dou- 31 ble up as informative podiums that catch real human 32 ∗ Corresponding author. A. Bhuvaneswari, Research Scholar, Department of Computer Technology, Madras Institute of Tech- nology, Anna University, Chennai. E-mail: bhuvana.cse14@ gmail.com. voices reaching out during disasters. However, the 33 information about ongoing events and sensitive issues 34 that remain unnoticed. Providentially, the CSN is 35 used as new additions to measure the impact of disas- 36 ter events due to wide usage of smart phones [5]. The 37 geographically linked CSN data containing geo-tags 38 is accepted as a trustworthy source for detecting dis- 39 aster and provides wide range of accessible services 40 [6]. In 2015, a study explored the retweeting activ- 41 ity geographically from collective Twitter population 42 around the globe during Hurricane Sandy [7] disas- 43 ter. They used feature-rich classifiers for determining 44 the microblogs which are most relevance to specified 45 event [8]. The events could be recognized as abnormal 46 spikes in activity [9] by monitoring the normal flow of 47 microblogs using change point detection approaches 48 [10]. It determines the intensity of the sensitive and 49 abnormal event that was deciphered in the course of 50 fluctuations in content. The annotation scheme [11] 51 for identifying relevant tweets after disasters usually 52 1064-1246/18/$35.00 © 2018 – IOS Press and the authors. All rights reserved