Abstract: Methods, apparatus, and systems directed to receiving search queries, retrieving documents, computing the number of categories to present for a given query, computing the number of results to show in each category, computing an ordering of categories, and for all the result pages beyond the first page employing user interface elements that optionally allow the user to quickly zoom in on a specific category and get more results belonging to that category.

Abstract: A tree structure has a node associated with each category of a hierarchy of item categories. Child nodes of the tree are associated with sub-categories of the categories associated with parent nodes. Training data including received queries and indicators of a selected item category for each received query is combined with the tree structure by associating each query with the node corresponding to the selected category of the query. When a query is received, a classifier is applied to the nodes to generate a probability that the query is intended to match an item of the category associated with the node. The classifier is applied until the probability is below a threshold. One or more categories associated with the nodes that are closest to the intent of the received query are selected and indicators of items of those categories that match the received query are output.

Abstract: A structured query may specify attribute values for attributes. An estimate of the number of items that will match the structured query if it is applied to a structured database is determined. If the estimated number of items is below a threshold, the structured query may be relaxed to form new candidate structured queries. The number of candidate queries may be determined based on a desired running time. Each of the candidate structured queries may be determined by changing one or more attribute values of the attributes of the structured query. Estimates of the number of items each of the candidate structured queries will match is determined, and the candidate structured query that has the highest matching estimation is used to query the database. The matching results may be output.

Abstract: The rank of nodes in a graph may be inferred from a calculated probability that each node in the graph appears in a single random walk of the graph. Short random walks may be generated for each node in the graph. The generated random walks may be combined to form a longer single random walk. Multiple single random walks may be generated in this fashion. The probability of each node appearing in a single random may be calculated from the generated single random walks. The rank of each node may then be inferred from the calculated probabilities.

Abstract: A document selector selects and ranks documents that are relevant to a query. The document selector executes an instance of a multi-armed bandits algorithm to select a document for each slot of a results page according to one or more strategies. The documents are selected in an order defined by the results page and documents selected for previous slots are used to guide the selection of a document for a current slot. If a document in a slot is subsequently selected, the strategy used to select the document is rewarded with positive feedback. When the uncertainty in an estimate of the utility of a strategy is less than the variation between documents associated with the strategy, the strategy is subdivided into multiple strategies. The document selector is able to “zoom in” on effective strategies and provide more relevant search results.

Abstract: Search results and ads that satisfy a query may be formatted into a search results page where an initial placement of ads may result in an estimated click through rate for the presented ads. An adjustment factor may be determined based on parameters associated with the search results, such as the clicks on the search results themselves. The adjustment factor may be applied to a particular ad to determine if an estimated click through rate of the particular ad will change with respect to a position when there are a first number of mainline ads and a second number of side bar ads. Mainline exclusivity may be appropriate for an ad to increase the click through rate of the ad. The increase may be determined in accordance with the adjustment factor to decide whether to present the ad with mainline exclusivity.

Abstract: Sketches are generated for each node in a graph. For undirected graphs, each sketch for a node may include an indicator of a node from a seed set of nodes and the shortest distance between the node and the indicated node. When a request is received for the shortest distance between two nodes of the graph, the sketches for each of the two nodes are retrieved, and nodes that are indicated in both of the sketches are determined. The distances between each of the two nodes and a determined node as indicated in the sketches is summed for each of the determined nodes, and the sum having the least distance is selected as the estimated shortest distance between the two nodes.

Abstract: Queries submitted by users looking for products and/or services are monitored and collected over a time period. Webpages corresponding to products and/or services bought by the users in response to submitting the queries are also monitored and collected over the time period. Attributes are extracted from the webpages and the queries, and the attributes are correlated to identify attributes that are similar to one another. The attributes are correlated to identify attributes that are not substitutable in a query. The identified attributes may be used to rank products and/or services that are responsive to a query based on attributes associated with the products and/or services, or to recommend alternative queries based on a submitted query by substituting one or more attributes of the query with similar attributes.

Abstract: Described is a search (e.g., web search) technology in which concepts are returned in response to a query in addition to (or instead of) search results in the form of traditional links. Each concept generally corresponds to a set of links to content that are more directed towards a possible user intention, or information need, with respect to that query. If a user selects a concept, that concept's links are exposed to facilitate selection of a document the user finds relevant. In this manner, much more than the top ten ranked links may be provided for a query, each set of other links arranged by the concepts. Also described is processing a query log or other data store to optionally find related queries and find the concepts, e.g., by clustering a relationship graph built from the query log to find dense subgraphs representative of the concepts.

Abstract: The rank of nodes in a graph may be inferred from a calculated probability that each node in the graph appears in a single random walk of the graph. Short random walks may be generated for each node in the graph. The generated random walks may be combined to form a longer single random walk. Multiple single random walks may be generated in this fashion. The probability of each node appearing in a single random may be calculated from the generated single random walks. The rank of each node may then be inferred from the calculated probabilities.

Abstract: As provided herein, a pairwise distance between nodes in a large graph can be determined efficiently. URL-sketches are generated for respective nodes in an index by extracting labels from respective nodes, which provide a reference to a link between the nodes, aggregating the labels into sets for respective nodes, and storing the sets of labels as URL-sketches. Neighborhood-sketches are generated for the respective nodes in the index using the URL-sketches, by determining a neighborhood for a node and generating a sketch using labels that are associated with the respective neighboring nodes. A distance between two nodes is determined by computing an approximate number of paths and an approximate path length between the two nodes, using the neighborhood sketches for the two nodes.

Abstract: A method is disclosed for identifying queries stored in a log which are semantically related to an input query that may include a large number of terms. A set of one or more subsequences are generated for each query stored in the log, and these sets of subsequences are stored in a lookup table. A set of one or more subsequences are also generated for the input query. The subsequences in the lookup table and of the input query are generated by hashing of the respective query terms to a value between 0 and 1 using a known technique of min-hashing. The present system then constructs the subsequences of the query using the k-min hashes of the query, where k is an integer based on the number of terms in the query.

Abstract: Techniques allow a client to be migrated from a session that has been established for a client on a server to another session on another server in a way that is a transparent to the application for which the initial session was established. The sessions are migrated under conditions in which a copy of session state of the session to migrate from the source node does not have to be captured and restored to a destination session.

Abstract: A system, method, computer program and system for suspending a result set and continuing from a suspended result set for a scrollable cursor is disclosed. The rows from a row source corresponding to the scrollable cursor result set are suspended into storage so that when a fetch is requested the data can be sent (continued) from the suspended storage and does not depend on the original scrollable cursor row source or session to be available.

Abstract: A model based on a generalization of the Examination Hypothesis is disclosed that states that for a given query, the user click probability on a document in a given position is proportional to the relevance of the document and a query specific position bias. Based on this model the relevance and position bias parameters are learned for different queries and documents. This is done by translating the model into a system of linear equations that can be solved to obtain the best fit relevance and position bias values. A cumulative analysis of the position bias curves may be performed for different queries to understand the nature of these curves for navigational and informational queries. In particular, the position bias parameter values may be computed for a large number of queries. Such an exercise reveals whether the query is informational or navigational.

Abstract: Methods, apparatus, and systems directed to receiving search queries, retrieving documents, computing the number of categories to present for a given query, computing the number of results to show in each category, computing an ordering of categories, and for all the result pages beyond the first page employing user interface elements that optionally allow the user to quickly zoom in on a specific category and get more results belonging to that category.

Abstract: Data from a click log may be used to generate training data for a search engine. The pages clicked as well as the pages skipped by a user may be used to assess the relevance of a page to a query. Labels for training data may be generated based on data from the click log. The labels may pertain to the relevance of a page to a query.

Abstract: A system, method, computer program and system for suspending a result set and continuing from a suspended result set for a cursor is disclosed. The rows from a row source corresponding to the cursor result set are suspended into storage so that when a fetch is requested the data can be sent (continued) from the suspended storage and does not depend on the original cursor row source or session to be available.

Abstract: A method and mechanism for implementing pipelined prefetching in a computer system is disclosed. Data prefetching is utilized to predictably retrieve information between multiple levels of nodes in the computer system, in which a first server may receive prefetched data from a second server, the second server receives prefetched data from a third server, with the chain of prefetching extending as long as needed/desired to effect efficient data transfers between the nodes.

Abstract: A summary of the neighborhood of a page may be determined offline and used at query time to approximate the neighborhood graph of the result set and to compute scores using the approximate neighborhood graph. The summary of the neighborhood graph may include a Bloom filter containing a limited size subset of ancestors or descendants of the page. A web page identifier may also be included in the summary. Consistent sampling is used, where a consistent unbiased sample of a number of elements from the set is determined. At query time, given a result set, the summaries for all the results may be used to create a cover set. An approximate neighborhood graph consisting of the vertices in the cover set is created. Ranking technique scores may be determined based on the approximate neighborhood graph.