<p> Social media is great resource of user-generated contents. Public attention, opinion and hot topics can be captured in the social media, which provides the ability to predict human related events. Since social media can be retrieved in real time with no building cost and no maintenance cost, traffic operation authorizes probably identify the social media data as another type sensor for traffic demand. In this thesis, we aim to use social media information to assist traffic flow prediction under special event conditions. Specially, a short-term traffic flow prediction model, incorporated with tweet features, is developed to forecast the incoming traffic flow prior sport game events. Both tweet rate features and semantic features are included in the prediction model. We examine and compare the performance of four regression methods, respectively autoregressive model, neural networks model, support vector regression, and k-nearest neighbor, with and without social media features. To the end, we show the benefit gained by including social media information in the prediction model and its computational efficiency for potential practical applications. </p>
| Identifer | oai:union.ndltd.org:PROQUEST/oai:pqdtoai.proquest.com:1546950 |
| Date | 07 December 2013 |
| Creators | Ni, Ming |
| Publisher | State University of New York at Buffalo |
| Source Sets | ProQuest.com |
| Language | English |
| Detected Language | English |
| Type | thesis |
Page generated in 0.1013 seconds