Vote Prediction on Comments in Social Polls Dataset
This page is a distribution site of social poll data for the task of Vote Prediction. Data available on this page include user demographic, comment, and vote information collected from the SodaHead social polling website.
The problem of Vote Prediction for which this dataset is intended is described in:
Social Poll Dataset
- Poll Data Includes question, answer, comment, and vote data from 4,803 polls (one poll per file).
- Users Includes anonymized demographic information about 108,462 users (one user per line).
- Cross Validation Folds Includes lists of the comment ids in each of the folds used for cross validation in the experiments described in Vote Prediction on Comments in Social Polls. Since we performed learning curve experiments, each file describes the cross-validation splitting at one of the four training set sizes used.
The creation of this website is based upon work supported in part by National Science Foundation (NSF) Grants IIS-1147644 and IIS-1219142. Any opinions, findings, and conclusions or
recommendations expressed above are those of the authors and do not
necessarily reflect the views of NSF and should
not be interpreted as representing the official policies, either expressed
or implied, of any sponsoring institution, the U.S. government or any other
If you have any questions or comments regarding this site, please send
email to Isaac Persing.