Predictive Discrete Latent Factor Models for Large Scale Dyadic Data  thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Predictive Discrete Latent Factor Models for Large Scale Dyadic Data

Published on Aug 15, 20077741 Views

We propose a novel statistical method to predict large scale dyadic response variables in the presence of covariate information. Our approach simultaneously incorporates the effect of covariates and e

Related categories

Chapter list

Predictive Discrete Latent Factor Models00:03
Internet Advertising: Billion Dollar Industry00:37
Recommender Systems01:26
Click Fraud01:57
Data02:39
Problem Definition04:01
Agenda05:15
Existing Approaches05:50
Non-Parametric Function Estimation06:03
Random Effects Model07:28
Generalized Linear Models09:25
Unsupervised Approach10:30
Clustering Animation11:09
PDLF: High Level Overview11:45
Model12:42
Fitting Algorithm: Generalized EM13:15
EM Algorithm Hard Clustering13:53
Simulation Study on Movie Lens 14:04
Logistic Regression on Movie Lens14:14
Experiments: Click Count Data14:27
Co-Cluster Interactions: Plain Co-Clustering15:02
Co-Cluster Interactions: PDLF15:49
Prediction Results16:35
Summary16:51
Ongoing Work17:36
Prediction Results (a)18:30