Discovery of unusual regional social activities using geo-tagged microblogs

Abstract

The advent of microblogging services represented by Twitter evidently stirred a popular trend of personal update sharing from all over the world. Furthermore, the recent mobile device and wireless network technologies are greatly expanding the connectivity between people over the social networking sites. Regarding the shared buzzes over the sites as a crowd-sourced database reflecting a various kind of real-world events, we are able to conduct a variety of social analytics using the crowd power in much easier ways. In this paper, we propose a geo-social event detection method by finding out unusually crowded places based on the conception of social networking sites as a social event detector. In order to detect unusual statuses of a region, we previously construct geographical regularities deduced from geo-tagged microblogs. Especially, we utilize a large number of geo-tagged Twitter messages which are collected by means of our own tweets acquisition method in terms of geographic relevancy. By comparing to those regularities, we decide if there are any unusual events happening in monitoring geographical areas. Finally, we describe the experimental results to evaluate the proposed unusuality detection method on the basis of geographical regularities which are computed from a large number of real geo-tagged tweet dataset around Japan.

Using Social Media for Emergency Response and Urban Sustainability: A Case Study of the 2012 Beijing Rainstorm

“…Combining texts with the location of twitters and visualizing the tag clouds on an interactive map, a scalable approach to detecting spatial cluster, Tom et al developed a new way to track and model abnormal events. Another example is one that used a clustering-based space partition method to detect normal and abnormal patterns as shown by geo-located twitter messages [14]. In this paper, we demonstrate a new way of applying spatial statistical analysis to analyze Sina-Weibo messages to study the spatial distribution of the incidents, even the anomalies.…”

“…Li [13] investigated the relationships between tweets and photo densities to explore the socioeconomic characteristics of creators by geographic data. To reflect a geographic region's status on Twitter, Lee et al [14] proposed an event detection system based on geographic regularity of the tweets. In addition, Cheng and Wicks [15] employed space-time scan statistics to detect statistically significant space-time events.…”

A Framework for Discovering Evolving Domain Related Spatio-Temporal Patterns in Twitter

“…Here, Twitter data is directly utilized to detect latent spatio-temporal clusters, outliers or hotspots without extracting topics, then deeper analysis is performed on any patterns detected to verify whether a special event has occurred. For example, Lee et al (2011) [10] divided the whole research region into sub-regions based on the spatial distribution of tweets by clustering. For each sub-region, the time stamps with unusually large number of tweets were then detected by boxplot.…”

“…Moreover, Twitter can be considered as a large black box that contains numerous topics reflecting various events from different domains, e.g., disasters [3], crimes [4], traffic [5], and epidemics [6]. Ways to extract hidden, unknown and significant events from the huge mass of Twitter data has thus become a research hotspot in computer science [7,8], human science [9,10] and GIS [11][12][13][14] in recent years. The research approaches applied can be roughly classified into three categories depending on which of the above three fields is the focus: (1) scholars in computer science consider tweets as textual information that changes over time, so topics related to different domains can be extracted by text classification methods such as Latent Dirichlet Allocation (LDA) and clustering; (2) in human science, scholars usually treat Twitter as a tool to record human behaviors; for example moving behaviors can be reflected by the changes in number of Twitter users coming into and going out of a certain region; and (3) researchers in GIS commonly extract domain related events to identify spatio-temporal outliers or hotspots.…”

Big Data Modeling and Analysis of Microblog Ecosystem

“…As Romero et al [17] proposed an passive-influence algorithm with consideration of both influence and passivity, we gave a formula to calculate users influence in SinaWeibo by considering users posts, followers, followings, and follow-back rate by referring to the research of Yang et al [18] who discovered that only 25.5% of all the information was generated by transfer via studying the transfer mechanism in Twitter. Users behavior and users intention were discussed in a variety of aspects, such as Lee et al [19] tried to mine users behavior pattern by marking users geography information via mobile microblog, while Lingad et al [20] used named entity recognizers to extract local information disaster-related microblogs and Kwak et al [21] discussed the dynamics of the behavior known as "unfollow" in Twitter and discovered the major factors affecting the decision to unfollow. We analyzed users intention by analyzing their microblog text and got interesting results of real big data of SinaWeibo with many first-hand new discoveries other than the small dataset results with traditional techniques, but natural language process (NLP) and big data analysis on the tags or semantic contents and contexts.…”