Abstract: In one embodiment, a method includes accessing an information graph that contains nodes and edges connecting the nodes. The nodes in the graph represent concepts related to at least one vertical. The method further includes associating a first node in the graph with a second node in the graph based on a relationship between the meaning of the concept represented by the first node and the meaning of the concept represented by the second node. A portion of the graph corresponds to an advertiser's product catalog. The method may further include accessing an ad library that contains elements for an advertisement. The method may further include selecting at least one of the elements based on the portion of the graph corresponding to the portion of the product catalog of the advertiser. The method may further include generating an advertisement for display based on the selected elements.

Abstract: In one embodiment, a method includes accessing an information graph that contains nodes and edges connecting the nodes. The nodes in the graph represent concepts related to at least one vertical. The method further includes associating a first node in the graph with a second node in the graph based on a relationship between the meaning of the concept represented by the first node and the meaning of the concept represented by the second node. A portion of the graph corresponds to an advertiser's product catalog. The method may further include accessing one or more targeting criteria for one or more products in the portion of the product catalog. The method may further include creating at least a portion of an advertising campaign, for at least one product for which targeting criteria has been accessed, based on the targeting criteria and the portion of the graph corresponding to the portion of the product catalog.

Abstract: In one embodiment, a method includes receiving, from an advertiser, an identification of one or more products in a catalog of the advertiser. The method may further include accessing an information graph that contains nodes and edges connecting the nodes. The nodes in the graph represent concepts related to at least one vertical. The method further includes associating a first node in the graph with a second node in the graph based on a relationship between the meaning of the concept represented by the first node and the meaning of the concept represented by the second node. A portion of the graph corresponds to an advertiser's product catalog. The method may further include identifying, based on the portion of the graph corresponding to the products identified by the advertiser, one or more products or product types related to the products identified by the advertiser.

Abstract: In one embodiment, a method includes creating an information graph that contains nodes and edges connecting the nodes. The nodes in the graph represent concepts related to at least one vertical. The method further includes associating a first node in the graph with a second node in the graph based on a relationship between the meaning of the concept represented by the first node and the meaning of the concept represented by the second node.

Abstract: In one embodiment, a method includes accessing information about an advertiser's product catalog. The method may further include accessing an information graph that contains nodes and edges connecting the nodes. The nodes in the graph represent concepts related to at least one vertical. The method further includes associating a first node in the graph with a second node in the graph based on a relationship between the meaning of the concept represented by the first node and the meaning of the concept represented by the second node. A portion of the graph corresponds to the advertiser's product catalog. The method may further include enriching information associated with the portion of the product catalog with information related to the concepts and relationships represented by the portion of the graph corresponding to the portion of the product catalog of the advertiser.

Abstract: The present invention is directed towards systems and method for generating a bid amount for one or more keywords for a display of an advertisement associated with a given keyword. The method comprises calculating estimates of one or more advertising metrics for one or more keywords associated with one or more advertisements displayed at one or more positions. Target positions are identified for the one or more keywords at which a return on investment earned by the display of one or more advertisements in response to a query that matches a given keyword exceeds a threshold by a given probability on the basis of the estimates to select a given target position that provides a greatest estimated profit. Bid amounts are generated for the one or more keywords in order to obtain the display of associated advertisements at the given target position.

Abstract: In one embodiment, a method includes accessing online advertising data including metric values for one or more online advertising spaces; identifying pairs of the user variables; for each of the pairs, calculating a statistical score for each of the online advertising metrics and calculating a weighted score based on all the statistical scores for the pair; selecting one of the pairs based on its weighted score; and performing a merge test on the selected pair; and merging the user variables in the selected one of the pairs with each other if the selected one of the pairs passes the merge test.

Abstract: In one embodiment, a method includes accessing metric values for one or more online advertising spaces for placing one or more online advertisements; generating a first user variable subgroup and a second user variable subgroup from a user variable group; deriving first metric values for the first user variable subgroup; deriving second metric values for the second user variable subgroup; for each of the online advertising metrics, using a particular statistical model to calculate a statistical score based on the first and second metric values with respect to the online advertising metric and the first and second user variable subgroups; performing a split test on the first user variable subgroup and the second user variable subgroup; and splitting the user variable group into two separate user variable groups if the first and second variable subgroups pass the split test.

Abstract: The present invention is directed towards systems and methods for generating bids for one or more keywords and advertisement groups. The method of the present invention comprises generating a data set with one or more entries, a given entry identifying one or more advertising metric values associated with an advertisement group and keyword. One or more keywords are selected from the one or more entries in the data set, and a bid is generated for the one or more selected keywords using the one or more advertising metric values associated with the one or more selected keywords. The one or more advertising metric values of the one or more entries associated with a given advertisement group exclusive of the one or more advertising metric values associated with the one or more selected keywords is aggregated. A bid is generated for the one or more advertisement groups using the one or more aggregated advertising metric values associated with the one or more advertisement groups.

Abstract: In particular embodiments, annotating each keyword of a plurality of keywords with one or more labels of a plurality of labels, comprising: for each label, computing a score for the keyword document corresponding to the keyword and the label using an annotation model; and annotating the keyword with a specific label where the keyword document corresponding to the keyword and the specific label have the highest or the lowest score. Constructing a classifier based on a plurality of training keywords. For each keyword of the plurality of keywords, for each label annotating the keyword, calculating a second index-wise product between a word count vector of the keyword document corresponding to the keyword and a word count vector of the label document corresponding to the label; and predicting whether the label annotating the keyword is correct using the classifier with the second index-wise product as an input to the classifier.

Abstract: In particular embodiments, constructing a classifier based on a plurality of training keywords. Each training keyword comprises one or more words and has a corresponding training keyword document. Constructing the classifier comprises: for each training keyword of the plurality of training keywords, annotating the training keyword with one or more labels of a plurality of labels, each label having a corresponding label document; and for each label annotating the training keyword, determining a correctness of the label; calculating a first index-wise product between a word count vector of the training keyword document corresponding to the training keyword and a word count vector of the label document corresponding to the label; and forming a pair of the correctness and the first index-wise product; and training the classifier using one or more pairs of the correctness and the first index-wise product.

Abstract: In one embodiment, a method includes accessing information regarding a visitor to a website; accessing a plurality of filters that are each associated with one of a plurality of lead buyers and being defined by a set of lead criteria specified by the associated lead buyer; and, for each of the filters, determining whether the information regarding the visitor satisfies the set of lead criteria defining the filter and, if the information regarding the visitor satisfies the set of lead criteria defining the filter, selecting the filter as a filter for which the visitor qualifies as a lead. The method includes, for each of the filters for which the visitor qualifies as a lead, accessing information indicating a potential revenue to be obtained by a lead matcher from the lead buyer associated with the filter.

Abstract: In one embodiment, a method includes deriving a user intent from user information associated with a user, selecting an advertiser intent that aligns with the user intent, and then advertising to the user based on the advertiser intent that aligns with the user intent.

Abstract: In one embodiment, a method includes constructing an intent map for a plurality of products, the intent map comprising intent topics and each intent topic comprising intents, and then deriving a plurality of keywords from the intent map based on keyword templates.

Abstract: In particular embodiments, advertising groups may be formed based on keywords. Access a plurality of keywords, each keyword comprising one or more words and describing one or more characteristics or features of particular advertising subject matter. Annotate each of the keywords with one or more labels selected from a plurality of labels, each label comprising one or more words and describing one or more aspects of or one or more categories or concepts represented by the keyword. Group the keywords into one or more keyword groups based on similarities between the labels of the keywords, each keyword group comprising one or more of the keywords. Form one or more advertising groups from the keyword groups, comprising from each of the keyword groups, form one or more of the advertising groups, each of the advertising groups comprising one or more of the keywords in the keyword group.

Abstract: In one example embodiment, a method includes receiving a request for content to be presented to a current user. The content has multiple elements that each have multiple possible element variations. The content has multiple possible instances that would result from different combinations of different ones of the possible element variations. The method includes accessing a current-user context of the current user that has data associated with or concerning the current user; predicting multiple responses of the current user to multiples ones of the possible instances of the content; selecting one of the possible instances of the content for the current user based on the predicted responses; and providing for presentation to the current user a current actual instance of the content based on the selected one of the possible instances of the content.

Abstract: In one embodiment, a method includes accessing a current-visitor context of a current visitor to a web page in a current web-browsing session. The current-visitor context includes one or more data associated with or concerning the current visitor. The method includes selecting based on the current-visitor context a particular one of multiple possible instances of the web page for presentation to the current visitor. The particular one of the multiple possible instances of the web page is substantially most likely to generate a highest expected outcome from interaction with the web page by the current visitor as indicated by the current-visitor context.

Abstract: Embodiments of a web page optimization system are described. The system utilizes an accumulated body of visitor data regarding interaction with one or more web pages to determine an optimized web page to display to a visitor. Various types of session data, including network traffic data for the client computer, demographic information about the visitor, and general and specific session data of other visitors in similar browsing sessions are defined and compiled. Continuous full-path optimization from initial visitor interaction with the web page is provided through a visitor context analysis component that analyzes certain parameters related to how the visitor has accessed the web page.

Abstract: Embodiments of a data representation system for describing specific data sets, such as documents, web pages, or search engine queries, based on data tokens, such as words or n-grams, contained in a collection of documents are described. Such a system can be used in any type of information retrieval application, such as a document, web page, or online advertisement serving process, based on an information request, such as a query executed through an Internet search engine. For example, when a search is performed at a search engine, a content provider uses the system to represent the search query and compares the query representation against representations of a set of content in order to identify, retrieve and aggregate the content from the set most relevant to the search query, in the form of a web page or other data unit for display or access through the web browser.

Abstract: The present invention is directed towards systems and method for generating a bid amount for one or more keywords for a display of an advertisement associated with a given keyword. The method comprises calculating estimates of one or more advertising metrics for one or more keywords associated with one or more advertisements displayed at one or more positions. Target positions are identified for the one or more keywords at which a return on investment earned by the display of one or more advertisements in response to a query that matches a given keyword exceeds a threshold by a given probability on the basis of the estimates to select a given target position that provides a greatest estimated profit. Bid amounts are generated for the one or more keywords in order to obtain the display of associated advertisements at the given target position.