200 likes | 365 Views
Local Event Locating and Mining on Twitter Data. Team 9: Karthik kumar Rangineni Zhi Liu. Event Locating and Mining on Twitter Data. Content. Reference Papers Introduction Papers Approach Experiment Conclusion. University of North Texas, Computer Science & Engineering.
E N D
Local Event Locating and Mining on Twitter Data Team 9: KarthikkumarRangineni Zhi Liu
Event Locating and Mining on Twitter Data Content • Reference Papers • Introduction • Papers • Approach • Experiment • Conclusion University of North Texas, Computer Science & Engineering
Geotagging Social Media for Enhanced Location-based Search Reference: • Sakaki, Takeshi, Makoto Okazaki, and Yutaka Matsuo, Earthquake shakes Twitter users: real-time event detection by social sensors., In proceedings of the 19th international conference on World wide web, pp. 851-860, ACM 2010 • Watanabe, Kazufumi, Masanao Ochi, Makoto Okabe, and RikioOnai, Jasmine: a real-time local-event detection system based on geolocation information propagated to microblogs., In proceedings of the 20th ACM international conference on Information and knowledge management, pp. 2541-2544, ACM, 2011 • Cheng, Zhiyuan, James Caverlee, and Kyumin Lee, You are where you tweet: a content-based approach to geo-locating twitter users., In proceedings of the 19th ACM international conference on Information and knowledge management, pp. 759-768, ACM 2010 University of North Texas, Computer Science & Engineering
Event Locating and Mining on Twitter Data Introduction • Event detection achieve more and more attention recently • Different event types: • Global events: trending, topics, news • Local events: show, ball games • Information used in event detection: • Context information • Time and location tag University of North Texas, Computer Science & Engineering
Event Locating and Mining on Twitter Data Introduction • Definition • Number of people participated • Time • Location • Subject University of North Texas, Computer Science & Engineering
Event Locating and Mining on Twitter Data Introduction • Why Twitter? University of North Texas, Computer Science & Engineering
Event Locating and Mining on Twitter Data Introduction • Large number of users University of North Texas, Computer Science & Engineering
Event Locating and Mining on Twitter Data Introduction • Large number of users • Real-time characteristic University of North Texas, Computer Science & Engineering
Event Locating and Mining on Twitter Data Introduction • Large number of users • Real-time characteristic • Time and location tag University of North Texas, Computer Science & Engineering
Event Locating and Mining on Twitter Data Approach • Event detection • Event time and location • Event description University of North Texas, Computer Science & Engineering
Event Locating and Mining on Twitter Data Approach – Paper 1 • Probabilistic Model • Temporal model • Spatial model Tweets related to earthquake Tweets related to typhoons 1. Earthquake shakes Twitter users: real-time event detection by social sensors University of North Texas, Computer Science & Engineering
Event Locating and Mining on Twitter Data Experiment and Examples of Event Detection System • Earthquake event detection University of North Texas, Computer Science & Engineering
Event Locating and Mining on Twitter Data Conclusion The authors investigated the real-time nature of Twitter, in particular for event detection. They considered each Twitter user as a sensor, and set a problem to detect an event based on sensory observations. Location estimation methods such as Kalmanfiltering and particle filtering are used to estimate the locations of events. As an application, the authors developed an earthquake reporting system, which is a novel approach to notify people promptly of an earthquake event. University of North Texas, Computer Science & Engineering
Event Locating and Mining on Twitter Data Approach – Paper 2 • Probabilistic Model • Temporal model • Spatial model • Geolocation information extract 2. Jasmine: a real-time local-event detection system based on geolocation information propagated to microblogs University of North Texas, Computer Science & Engineering
Event Locating and Mining on Twitter Data Experiment and Examples of Event Detection System • Jasmine University of North Texas, Computer Science & Engineering
Event Locating and Mining on Twitter Data Conclusion The authors proposed and evaluate a probabilistic framework for estimating a Twitter user's city-level location based purely on the content of the user's tweets, even in the absence of any other geospatial cues. By augmenting the massive human-powered sensing capabilities of Twitter and related microbloggingservices with content-derived location information, this framework can overcome the sparsity of geo-enabled features in these services and enable new location based personalized information services, the targeting of regional advertisements, and so on University of North Texas, Computer Science & Engineering
Event Locating and Mining on Twitter Data Approach – Paper 3 • Probabilistic Model • Temporal model • Spatial model • Geolocation information extract • User locating by context information 3. You are where you tweet: a content-based approach to geo-locating twitter users University of North Texas, Computer Science & Engineering
Event Locating and Mining on Twitter Data Experiment and Examples of Event Detection System • User Locating University of North Texas, Computer Science & Engineering
Event Locating and Mining on Twitter Data Conclusion Jasmine is a real-time local-event detection system. The authors demonstrate that this automatic geotagging method successfully assigned location information to non-geotaggedTwitter documents and increases the number of popular places detected by about 115 times. This system extracted key terms that describe the details of a local event. In a subjective evaluation, this system detected meaningful real-world local events with an accuracy of 25.5 %. To the future, the authors want to assign location information to many more non-geotagged Twitter documents by analyzing histories of users' movements and textual contexts. University of North Texas, Computer Science & Engineering
Geotagging Social Media for Enhanced Location-based Search Thanks! University of North Texas, Computer Science & Engineering