Call for Challenge
This challenge is entity type prediction over linked data. In this challenge, 1897 URLs are provided and 1397 of them are provided with label information. The task is the classification of the 500 unlabeled URLs. The linked data about all the 1897 URLs are also provided.
Train/Test Data and Evaluation are described in PDF version. CallForChallenge.pdf
Each submission include: 1) the result file, and 2) the associated documentation with environment setting and algorithm description. The above materials should be sent to both email@example.com and before the deadline, and the email title should be 'teamname+datachallenge'.
The result file should be named "result.dat", and the format must be the same as that of train.dat, i.e. each line contains two columns, URL and the label, and the two columns are separated by tab.
The associated documentation should be named “datachallenge.pdf” and provided in the PDF format, using the style of the Springer Publications format for Lecture Notes in Computer Science (LNCS). The document must be no longer than *5* pages.
If you have any questions, please contact .
Nov 1, 2015
Nov 4, 2015
The top 3 teams may be invited with free registration to present their solutions in the meeting.