Discovering Spatio-temporal Patterns of Themes in Social Media

Social networking website creates new ways for engaging people belonging to different communities, moral and social values to communicate and share valuable knowledge, therefore creating a large amount of data. The importance of mining social media cannot be over emphasized, due to significant information that are revealed which can be applied in different areas. In this paper, a systematic approach for traversing the content of weblog, considering location and time (spatiotemporal) is proposed. The proposed model is capable of searching for subjects in social media using Boyer Moore Horspool (BMH) algorithm with respect to location and time. BMH is an efficient string searching algorithm, where the search is done in such a way that every character in the text needs not to be checked and some characters can be skipped without missing the subject occurrence. Semantic analysis was carried out on the subject by computing the mean occurrence of the subject with the corresponding predicate and object from the total occurrence of the subject. Experiments were carried out on two datasets: the first category was crawled from twitter website from September to October 2014 and the second category was obtained from spinn3r dataset made available through the International AAAI (Association for the Advancement of Artificial Intelligence) Conference on Web and Social Media (ICWSM). The results obtained from tracking some subjects such as Islam and Obama shows that the mean occurrence of the analysis of the subject successfully reveals the pattern of the subject over a period of time for a specific location. Evaluation of the system which is based on performance and functionality reveals that the model performs better than some baseline models. The proposed model is capable of revealing spatiotemporal pattern for a subject, and can be applied in any area where spatiotemporal factor is to be considered. keywords: Boyer-Moore-Horspool Alogrithm, Search processing, Spatio temporal pattern, Sementic analysis.

[1]  Bing Liu,et al.  Sentiment Analysis and Opinion Mining , 2012, Synthesis Lectures on Human Language Technologies.

[2]  Lemi Baruh,et al.  Conference Report: The 6th International AAAI Conference on Weblogs and Social Media, June 4-7, 2012 , 2012, Int. J. Interact. Commun. Syst. Technol..

[3]  Michael I. Jordan,et al.  Latent Dirichlet Allocation , 2001, J. Mach. Learn. Res..

[4]  Charu C. Aggarwal,et al.  Text Mining in Social Networks , 2011, Social Network Data Analytics.

[5]  Bill Tomlinson,et al.  America is like Metamucil: fostering critical and creative thinking about metaphor in political blogs , 2010, CHI.

[6]  Jayanta Kumar Pal,et al.  Identifying Themes in Social Media and Detecting Sentiments , 2012 .

[7]  Bolanle Adefowoke Ojokoh,et al.  Generating Recommendation Status of Electronic Products from Online Reviews , 2013 .

[8]  Stuart E. Middleton,et al.  Social media text mining and network analysis for decision support in natural crisis management , 2013, ISCRAM.

[9]  Ravi Kumar,et al.  Structure and evolution of blogspace , 2004, CACM.

[10]  Chong Wang,et al.  Mining geographic knowledge using location aware topic model , 2007, GIR '07.

[11]  Stuart E. Middleton,et al.  Social-media text mining and network analysis to support decision support for natural crisis management , 2013 .

[12]  Andrew Hume,et al.  Fast string searching , 1991, USENIX Summer.

[13]  S. Renukappa,et al.  Examining the Satisfaction Level of Construction Workers Regarding Safety Management in the Kingdom of Saudi Arabia , 2017 .

[14]  William M. Campbell,et al.  Social Network Analysis with Content and Graphs , 2013 .

[15]  Andrew McCallum,et al.  Topics over time: a non-Markov continuous-time model of topical trends , 2006, KDD '06.

[16]  Markus E. Nebel.  Searching Texts - But Fast! The Boyer-Moore-Horspool Algorithm , 2011, Algorithms Unplugged.

[17]  Robert S. Boyer,et al.  A fast string searching algorithm , 1977, CACM.

[18]  Matthew Hurst,et al.  Event Detection and Tracking in Social Streams , 2009, ICWSM.

[19]  Maxime Crochemore,et al.  Boyer-Moore Strategy to Efficient Approximate String Matching , 1996, CPM.

[20]  Susanne Heuser,et al.  Location Based Social Networks – Definition, Current State of the Art and Research Agenda , 2013, Trans. GIS.

[21]  Divyakant Agrawal,et al.  Structural Trend Analysis for Online Social Networks , 2011, Proc. VLDB Endow..

[22]  Chao Liu,et al.  A probabilistic approach to spatiotemporal theme pattern mining on weblogs , 2006, WWW '06.

[23]  Kalev Leetaru,et al.  Culturomics 2.0: Forecasting large-scale human behavior using global news media tone in time and space , 2011, First Monday.