Abstract: A system comprises users disposed in various locations within a region, each user associated with an advertisement of an associated web property, each advertisement for enabling visitors to navigate to the associated web property; pods distributed within the region, each user being associated with at least one pod located more proximate to the user than another pod and that stores account information on the advertisement associated with the user; an advertisement channel for receiving a content request from a visitor, for obtaining an advertisement of a web property based on the content request, and for presenting the advertisement to the visitor in response to the content request; and a tag-based tracking mechanism for collecting visitor events if the visitor navigates to the web property and for forwarding visitor data based on the visitor events to the pod associated with the user that is associated with the advertisement.

Abstract: Generally, the present invention provides a method and computerized system for generating a classifier model, wherein the classifier model is operative to classify web content. The method and computerized system includes a first step of defining a plurality of predictive performance measures based on a leave one out (LOO) cross validation in terms of selectable model parameters. Exemplary predictive performance measures includes smoothened predictive measures such as F-measure, weighted error rate measure, area under curve measure, by way of example. The method and computerized system further includes deriving efficient analytical expressions for predictive performance measures to compute the LOO predictive performance and their derivatives. The next step is thereupon selecting a classifier model based on the LOO predictive performance.

Abstract: The present invention relates to systems and methods for the optimized selection and delivery of one or more advertisements from among one or more advertising campaigns. The method of the present invention comprises generating one or more media plans identifying execution parameters for the optimized selection and delivery of one or more advertisements. One or more advertisements organized according to one or more advertisement campaigns are retrieved. Additionally, advertiser specified constraint and target values associated with the one or more advertisements are retrieved. A forecast for the performance of the one or more advertisements is generated. A media plan is generated for the one or more advertisements according to the constraint and target values, as well as the forecast data. The one or more advertisements are distributed according to the execution parameters identified by the media plan.

Abstract: Generally, the present invention provides for monitoring user Internet activity based on DNS information. The invention includes accessing a plurality of query logs from a plurality of DNS servers and retrieving a plurality of DNS requests from the query logs. The DNS requests are requests from a various computing devices that have requested web content from web servers indicated by IP addresses. The invention further includes processing the DNS requests to determine Internet-based activity information by the plurality of computing devices that requested the web content and determining attention data based on the activity information. The attention data indicates web search trends based on the corresponding IP addresses accessed by the requesting computing devices.