This Title All WIREs
How to cite this WIREs title:
WIREs Data Mining Knowl Discov
Impact Factor: 1.939

# An overview on the exploitation of time in collaborative filtering

Can't access this content? Tell your librarian.

Classic Collaborative Filtering (CF) algorithms rely on the assumption that data are static and we usually disregard the temporal effects in natural user‐generated data. These temporal effects include user preference drifts and shifts, seasonal effects, inclusion of new users, and items entering the system—and old ones leaving—user and item activity rate fluctuations and other similar time‐related phenomena. These phenomena continuously change the underlying relations between users and items that recommendation algorithms essentially try to capture. In the past few years, a new generation of CF algorithms has emerged, using the time dimension as a key factor to improve recommendation models. In this overview, we present a comprehensive analysis of these algorithms and identify important challenges to be faced in the near future. WIREs Data Mining Knowl Discov 2015, 5:195–215. doi: 10.1002/widm.1160

• Algorithmic Development > Spatial and Temporal Data Mining
• Application Areas > Data Mining Software Tools
Example of feedback matrices: (a) a typical numerical ratings matrix and (b) a positive‐only ratings matrix.
[ Normal View | Magnified View ]
CANDECOMP/PARAFAC tensor factorization model: three matrices are obtained covering the same latent feature space f1 … k. The time‐related dimension t1tl is a time feature extracted from the ratings timestamp—e.g., day of week, month, and hour of day. Every cell in the original tensor can be predicted using the inner product of the three corresponding vectors $R^uit=∑d=1kUudIidTtd$.
[ Normal View | Magnified View ]

### Browse by Topic

Algorithmic Development > Spatial and Temporal Data Mining
Application Areas > Data Mining Software Tools