This item is a Paper in the Data Mining track.
- Agarwal, Deepak - Yahoo! Laboratories
- Chen, Bee-Chung - Yahoo! Laboratories
- Elango, Pradheep - Yahoo! Laboratories
We propose novel spatio-temporal models to estimate clickthrough rates in the context of content recommendation. We track article CTR at a ﬁxed location over time through a dynamic Gamma-Poisson model and combine information from correlated locations through dynamic linear regressions, signiﬁcantly improving on per-location model. Our models adjust for user fatigue through an exponential tilt to the ﬁrstview CTR (probability of click on ﬁrst article exposure) that is based only on user-speciﬁc repeat-exposure features. We illustrate our approach on data obtained from a module (Today Module) published regularly on Yahoo! Front Page and demonstrate signiﬁcant improvement over commonly used baseline methods. Large scale simulation experiments to study the performance of our models under different scenarios provide encouraging results. Throughout, all modeling assumptions are validated via rigorous exploratory data analysis.
Export Record As...
- HTML Citation
- ASCII Citation
- Resource Map
- OpenURL ContextObject
- OpenURL ContextObject in Span
- EP3 XML
- Dublin Core
- Reference Manager
- Eprints Application Profile
- Simple Metadata