0.25
0.5
0.75
1.25
1.5
1.75
2
Large Scale Ranking Problem: some theoretical and algorithmic issues
Published on Feb 25, 20076417 Views
The talk is divided into two parts. The first part focuses on web-search ranking, for which I discuss training relevance models based on DCG (discounted cumulated gain) optimization. Under this metric
Related categories
Chapter list
Advertizing00:05
Ranking Problems02:04
Earlier Work on Statistical Ranking04:06
Theoretical Results on Ranking07:38
Web-Search Problem09:39
Relevance Ranking: Statistical Learning Formulation12:32
Measuring Ranking Quality13:49
Subset Ranking Model15:23
Some Theoretical Questions16:40
Bayes Optimal Scoring17:42
Simple Regression18:36
Importance Weighted Regression20:38
Relationship of Regression and Ranking22:11
Appropriate Parameter Choice (for previous Theorem)22:58
Generalization Performance with Square Regularization23:03
Interpretation of Results23:41
Another Ranking Example: spelling correction in<br> web-search24:55
Some Conclusions25:49