en
0.25
0.5
0.75
1.25
1.5
1.75
2
A Hazard Based Approach to User Return Time Prediction
Published on Oct 07, 20141785 Views
In the competitive environment of the internet, retaining and growing one's user base is of major concern to most web services. Furthermore, the economic model of many web services is allowing free ac
Related categories
Chapter list
A Hazard Based Approach to User Return Time Prediction00:00
Outline00:18
Motivation00:22
User Retention - 100:23
User Retention - 200:32
User Retention - 300:44
Retention for the New Web - 100:56
Retention for the New Web - 202:15
Model the Time of Next Return02:19
Techniques02:37
Survival Analysis02:39
Cox Proportional Hazard Model - 103:58
Cox Proportional Hazard Model - 204:21
Cox Proportional Hazard Model - 304:48
Cox Proportional Hazard Model - 405:18
Cox Proportional Hazard Model - 505:25
Cox Proportional Hazard Model - 605:43
Cox Proportional Hazard Model - 705:53
Cox Proportional Hazard Model - 806:03
Data and Findings06:17
Music Domain Datasets06:20
Covariates07:04
Baseline Hazard Function08:09
Return Time Prediction - 108:32
Return Time Prediction - 208:35
Future Return Time Prediction - 108:55
Future Return Time Prediction - 209:23
Classification into User Buckets09:58
F-measure for User Classification using LOA (Large-scale Dataset)10:09
Conclusions10:33
Takeaways10:35
Thanks!11:14