Digging into Data: Data Mining II

Dionysios Kehagius of the Centre for Research and Technology Hellas presents the effects of global metrics in travel time forecasting.

After Matthew Karlaftis’ morning speech highlighting the importance of the use of big data, it is no surprise that many of the day’s sessions focused on the use of data and data mining. The creatively titled “Data Mining II” session featured several presentations about the use of data. Presentations in this session focused on GPS data use as well as roadside sensors.

 

 

More detailed data can be used to aid in travel time predictions, traffic predictions, route finding assistance, and transpor model development. GPS data was used to estimate lane level road data by Champike Uduwaragoda (Univ. or Moratuwa), to accurately detect travel mode by Richard Brunauer (Salzburg Res. Forschungsgesellschaft mbH), and for travel time forecasting by Dionysios Kehagias (Centre for Research and Technology Hellas). Data from roadside sensors was used to predict unusual events, such as crashes, by Konstantinos Kalpakis.

The abundance of GPS data in our world means that we can get more precise answers to old questions and start to answer new questions. However, one of the major challenges is access to this data for researchers. This often can prevent the generalization of results that would be a logical next step in many of these projects. Said Dr. Kehagias in his closing remark “I wish I had more data.”

Be Sociable, Share!
This entry was posted in Uncategorized. Bookmark the permalink.