Abstract
We describe the Joint Effort-Topic (JET) model and the Author Joint Effort-Topic (aJET) model that estimate the effort required for users to contribute on different topics. We propose to learn word-level effort taking into account term preference over time and use it to set the priors of our models. Since there is no gold standard which can be easily built, we evaluate them by measuring their abilities to validate expected behaviours such as correlations between user contributions and the associated effort.
Original language | English |
---|---|
Title of host publication | WWW 2014 Companion - Proceedings of the 23rd International Conference on World Wide Web |
Place of Publication | New York, NY (US) |
Publisher | ACM |
Pages | 549-550 |
Number of pages | 2 |
ISBN (Print) | 978-1-4503-2745-9 |
DOIs | |
Publication status | Published - 7 Apr 2014 |
Event | 23rd International Conference on World Wide Web - Seoul, Korea, Republic of Duration: 7 Apr 2014 → 11 Apr 2014 |
Conference
Conference | 23rd International Conference on World Wide Web |
---|---|
Abbreviated title | WWW 2014 |
Country/Territory | Korea, Republic of |
City | Seoul |
Period | 7/04/14 → 11/04/14 |