Google Inc.

University of Minnesota

Levent Ertöz

Marc Berndl

Sridhar Ramaswamy

Myles A Sussman

Adam Isaac Juda

Vipin Kumar

Nicholas Charles Fox

Karthik Gopalratnam

Dan Liu

Jonathan G Alferness

In general, in one aspect, a first request to provide one or more advertisements on a first web page is received, the first web page comprising one or more items of web content related to a query. A first quality score is calculated for a first advertisement included in a set of one or more candidate advertisements. The first advertisement is presented on the first web page based at least in part on the first quality score. A second request is received to provide one or more advertisements on a second web page, the second web page comprising one or more different items of web content related to the query. A second quality score for the first advertisement is calculated based at least in part on the previous presentation of the first advertisement. It is determined whether to present the first advertisement on the second web page based at least in part on the second quality score.

Advertisement selection

Identifying anomalies or outliers in a set of data records employs a distance or similarity measure between features of record pairs that depends upon the frequencies of the feature values in the set. Feature distances may be combined for a total distance between record pairs. An outlier is indicated for a certain score that may be based upon the pairwise distances. Outliers may be employed to detect intrusions in computer networks.

Identification of anomalous data records

Growing community of inventors

Mountain View, CA, United States of America

Levent Ertöz