Abstract: Methods and systems are provided for click through rate prediction and advertisement selection in online advertising. Methods are provided in which output information from a feature-based machine learning model is utilized. The output information includes predicted click through rate information. The output information is used to form a matrix. The matrix is modeled using a latent variable model. Machine learning techniques can be used in determining values for unfilled cells of one or more model matrices. The latent variable model can be used in determining predicted click through rate information, and in advertisement selection in connection with serving opportunities.

Abstract: A method includes describing the thread configurations of a volume of well-ordered electronic message transmissions (EMT) and utilizing the thread configuration data to conduct selective searches of the EMT volume. An apparatus includes a thread processor and a query manager. The thread processor analyzes the EMT threads and records the thread configuration data. The query manager utilizes the thread configuration data to conduct selective searches of the EMT volume.

Abstract: Systems and methods for building a prediction model to predict a degree of relevance between digital ads and a search query or webpage content are disclosed. Generally, an indication of relevance is received between a plurality of digital ads and one of a webpage content or a search query. A set of features is extracted from the plurality of digital ads and one of the webpage content or the search query. A prediction model is then built to predict a degree of relevance between the set of candidate digital ads and one of a second webpage content or a second search query, where the prediction model is built based at least one the received indication of relevance and the extracted set of features.

Abstract: Systems and methods for predicting a degree of relevance between a set of candidate digital ads and webpage content are disclosed. Generally, an ad provider receives a digital ad request associated with webpage content. The ad provider identifies a set of candidate digital ads that may be served in response to the digital ad request. A relevance module extracts a set of features from the set of candidate digital ads and the webpage content, and determines a degree of relevance between the set of candidate digital ads and the webpage content based on a prediction model and the extracted set of features. If the relevance module determines the set of candidate digital ads is relevant to the webpage content, the ad provider may serve one or more digital ads from the set of candidate digital ads in response to the received digital ad request.

Abstract: Systems and methods for predicting a degree of relevance between a set of candidate digital ads and a search query are disclosed. Generally, an ad provider receives a digital ad request associated with a search query. The ad provider identifies a set of candidate digital ads that may be served in response to the digital ad request. A relevance module extracts a set of features from the set of candidate digital ads and the search query associated with the digital ad request, and determines a degree of relevance between the set of candidate digital ads and the search query based on a prediction model and the extracted set of features. If the relevance module determines the set of candidate digital ads is relevant to the search query, the ad provider may serve one or more digital ads from the set of candidate digital ads in response to the received digital ad request.

Abstract: A method is provided to match an advertisement to a search query comprising: receiving search results produced by a search engine in response to a search query; producing an ad query that includes, unigram features, classification features with respect to an external classification system, and phrase features; producing a plurality of representations of corresponding advertisements in terms of the same types of features; and selecting one or more advertisements based upon a measure of similarity of ad query features to advertisements represented in terms of the same features.

Abstract: A method includes describing the thread configurations of a volume of well-ordered electronic message transmissions (EMT) and utilizing the thread configuration data to conduct selective searches of the EMT volume. An apparatus includes a thread processor and a query manager. The thread processor analyzes the EMT threads and records the thread configuration data. The query manager utilizes the thread configuration data to conduct selective searches of the EMT volume.

Abstract: Systems and methods for estimating an amount of traffic associated with a digital ad are disclosed. Generally, a forecasting module identifies a set of candidate webpages on which a digital ad may be displayed and estimates a click through rate associated with the digital ad and a webpage of the set of candidate webpages. The forecasting module determines a ranking score associated with the digital ad based on the determined click through rate and a bid price associated with the digital ad. The forecasting module then examines historical data, such as search logs, to determine an estimate of traffic associated with the digital ad with respect to the webpage in response to determining the ranking score of the digital ad exceeds a ranking score associated with another digital ad that was previously displayed on the webpage.

Abstract: A method for querying multifaceted information. An inverted index is constructed to include unique indexed tokens associated with posting lists of one or more documents. An indexed token is either a facet token included in a document as an annotation or a path prefix of the facet token. The annotation indicates a path within a tree structure representing a facet that includes the document. The tree structure includes nodes representing categories of documents. Constructing the inverted index includes generating a full path token and an associated full path token posting list. A query is received that includes constraints on documents. The constraints are associated with indexed tokens and corresponding posting lists. An execution of the query includes identifying the corresponding posting lists by utilizing the constraints and the inverted index and intersecting the posting lists to obtain a query result.

Abstract: A method and system for querying multifaceted information. An inverted index is constructed to include unique indexed tokens associated with posting lists of one or more documents. An indexed token is either a facet token included in a document as an annotation or a path prefix of the facet token. The annotation indicates a path within a tree structure representing a facet that includes the document. The tree structure includes nodes representing categories of documents. A query is received that includes constraints on documents. The constraints are associated with indexed tokens and corresponding posting lists. An execution of the query includes identifying the corresponding posting lists by utilizing the constraints and the inverted index and intersecting the posting lists to obtain a query result.

Abstract: A method for querying multifaceted information. An inverted index is constructed to include unique indexed tokens associated with posting lists of one or more documents. An indexed token is either a facet token included in a document as an annotation or a path prefix of the facet token. The annotation indicates a path within a tree structure representing a facet that includes the document. The tree structure includes nodes representing categories of documents. Constructing the inverted index includes generating a full path token and an associated full path token posting list. A query is received that includes constraints on documents. The constraints are associated with indexed tokens and corresponding posting lists. An execution of the query includes identifying the corresponding posting lists by utilizing the constraints and the inverted index and intersecting the posting lists to obtain a query result.

Abstract: Disclosed is a technique for building an index. A new indexi+1 is built and an anchor text tablei+1 and a duplicates tableti+1 are output using a storesi, a delta store, and previously generated global analysis computationsi, wherein the previously generated global analysis computationsi include an anchor text tablei, a rank tablei, and a duplicates tablei. New global analysis computationsi+1 are generated using the anchor text tablei+1, the duplicates tablei+1, and the previously generated global analysis computationsi.

Abstract: Disclosed is a technique for indexing data. For each token in a set of documents, a sort key is generated that includes a document identifier that indicates whether a section of a document associated with the sort key is an anchor text section or a context section, wherein the anchor text section and the context text section have a same document identifier; it is determined whether a data field associated with the token is a fixed width; when the data field is a fixed width, the token is designated as one for which fixed width sort is to be performed; and, when the data field is a variable length, the token is designated as one for which a variable width sort is to be performed. The fixed width sort and the variable width sort are performed. For each document, the sort keys are used to bring together the anchor text section and the context section of that document.

Abstract: A method that eagerly evaluates predicates of XPath queries over XML document nodes for a set of commonly known functions and operators (including arithmetic, general comparison, value comparison, Boolean operators, etc.) without materializing sequences is discussed. Such eager evaluation of predicates reduces the amount of buffer space required since evaluation sequences have to be buffered only partially during the predicate evaluation process. Document nodes to be selected by a query are determined earlier so that they can be outputted without buffering.

Abstract: A system, method, and computer program product to improve XML query processing efficiency with virtual cursors. Structural joins are a fundamental operation in XML query processing, and substantial work exists on index-based algorithms for executing them. Two well-known index features—path indices and ancestor information—are combined in a novel way to replace at least some of the physical index cursors in a structural join with virtual cursors. The position of a virtual cursor is derived from the path and ancestor information of a physical cursor. Virtual cursors can be easily incorporated into existing structural join algorithms. By eliminating index I/O and the processing cost of handling physical inverted lists, virtual cursors can improve the performance of holistic path queries by an order of magnitude or more.

Abstract: A method includes describing the thread configurations of a volume of well-ordered electronic message transmissions (EMT) and utilizing the thread configuration data to conduct selective searches of the EMT volume. An apparatus includes a thread processor and a query manager. The thread processor analyzes the EMT threads and records the thread configuration data. The query manager utilizes the thread configuration data to conduct selective searches of the EMT volume.

Abstract: Provided are techniques for computer-based electronic Information Retrieval (IR). An extended inverted index structure by generating one or more lexical affinities (LA), wherein each of the one or more lexical affinities comprises two or more search items found in proximity in one or more documents in a pool of documents, and generating a posting list for each of the one or more lexical affinities, wherein each posting list is associated with a specific lexical affinity and contains document identifying information for each of the one or more documents in the pool that contains the specific lexical affinity and a location within the document where the specific lexical affinity occurs.

Abstract: A method for indexing a plurality of documents, that includes a plurality of duplicate documents, first identifies one or more duplicate groups of documents from among the plurality of documents. Then, one index of content for the duplicate group is created instead of indexing the content from every document within the duplicate group. However, in contrast to the content index, an index of metadata for each of the documents in the duplicate group is created. Thus the content of each duplicate group is indexed only once, while a search engine using such indexing techniques retains the capability to answer queries as if the duplicated content was indexed for each document of the group.

Abstract: Provided are a method, system, and program for searching documents for ranges of numeric values. Document identifiers for documents are accessed, wherein the documents include at least one value that is a member of a set of values. A number of posting lists are generated. Each posting list is associated with a range of consecutive values within the set of values and includes document identifiers for documents having values within the range of consecutive values associated with the posting list. Each document identifier is associated with one value in the set of values included in the document identified by the document identifier. The generated posting lists are stored.

Abstract: Disclosed is a technique for indexing data. A token is received. It is determined whether a data field associated with the token is a fixed width. When the data field is a fixed width, the token is designated as one for which fixed width sort is to be performed. When the data field is a variable length, the token is designated as one for which a variable width sort is to be performed.