Abstract: Some implementations generate a mapping function using one or more historic performance indicators for a set of ad-keyword pairs and one or more advertisement metrics extracted from the set of ad-keyword pairs. The mapping function may be applied to map one or more advertisement metrics of a particular ad-keyword pair to determine a quality score for the particular ad-keyword pair. For example, the quality score may be used when determining whether to select an advertisement for display or may be provided as feedback to an advertiser. Additionally, in some implementations, the mapping function may be applied to determine a quality score for a new ad-keyword pair that has not yet accumulated historic information.

Abstract: Some implementations construct a quality score table based on historic data collected for a plurality of ad-keyword pairs. An ad-keyword pair may be selected for determining a quality score. One or more advertisement parameters may be determined for the selected ad-keyword pair. Based on the one or more advertisement parameters, the quality score for the selected ad-keyword pair may be determined from the quality score table. In some implementations, the quality score table is constructed by iteratively cutting a directed graph representing the advertisement parameters and the historic data. Further, in some implementations, the table may be smoothed using a smoothing operation.

Abstract: Ranking documents based on a series of web graphs collected over time is provided. A ranking system provides multiple transition probability distributions representing different snapshots or times. Each transition probability distribution represents a probability of transitioning from one document to another document within a collection of documents using a link of the document. The ranking system determines a stationary probability distribution for each snapshot based on the transition probability distributions for that snapshot and the stationary probability distribution of the previous snapshot. The stationary probability distributions represent a ranking of the documents over time.

Abstract: Embodiments for caching and accessing Directed Acyclic Graph (DAG) data to and from a computing device of a DAG distributed execution engine during the processing of an iterative algorithm. In accordance with one embodiment, a method includes processing a first subgraph of the plurality of subgraphs from the distributed storage system in the computing device. The first subgraph being processed with associated input values in the computing device to generate first output values in an iteration. The method further includes storing a second subgraph in a cache of the device. The second subgraph being a duplicate of the first subgraph. Moreover, the method also includes processing the second subgraph with the first output values to generate second output values if the device is to process the first subgraph in each of one or more subsequent iterations.

Abstract: Systems, methods, and devices for sorting and processing various types of graph data are described herein. Partitioning graph data into master data and associated slave data allows for sorting of the graph data by sorting the master data. In another embodiment, promoting a data bucket having a first data bucket size to a data bucket having a second data bucket size greater than the first data bucket size upon reaching a memory limit allows for the reduction of temporary files output by the data bucket.

Abstract: An anti-spam technique for protecting search engine ranking is based on mining search engine optimization (SEO) forums. The anti-spam technique collects webpages such as SEO forum posts from a list of suspect spam websites, and extracts suspicious link exchange URLs and corresponding link formation from the collected webpages. A search engine ranking penalty is then applied to the suspicious link exchange URLs. The penalty is at least partially determined by the link information associated with the respective suspicious link exchange URL. To detect more suspicious link exchange URLs, the technique may propagate one or more levels from a seed set of suspicious link exchange URLs generated by mining SEO forums.

Abstract: Some implementations provide techniques for determining which URLs to select for crawling from a pool of URLs. For example, the selection of URLs for crawling may be made based on maintaining a high coverage of the known URLs and/or high discoverability of the World Wide Web. Some implementations provide a multi-level coverage strategy for crawling selection. Further, some implementations provide techniques for discovering unseen URLs.

Abstract: Some implementations provide techniques for selecting web pages for inclusion in an index. For example, some implementations apply regularization to select a subset of the crawled web pages for indexing based on link relationships between the crawled web pages, features extracted from the crawled web pages, and user behavior information determined for at least some of the crawled web pages. Further, in some implementations, the user behavior information may be used to sort a training set of crawled web pages into a plurality of labeled groups. The labeled groups may be represented in a directed graph that indicates relative priorities for being selected for indexing.

Abstract: The present invention provides an improved method for ranking documents using a ranking model. One embodiment employs Continuous Conditional Random Fields (CRF) as a model, which is a conditional probability distribution representing a mapping relationship from retrieved documents to their ranking scores. The model can naturally utilize features of the content information of documents as well as the relation information between documents for global ranking. The present invention also provides a learning algorithm for creating Continuous CRF. Also provided, the invention introduces Pseudo Relevance Feedback and Topic Distillation.

Abstract: An importance system calculates the importance of pages using a conditional Markov random walk model rather than a conventional Markov random walk model. The importance system calculates the importance of pages factoring in the importance of sites that contain those pages. The importance system may factor in the importance of sites based on the strength of the correlation of the importance of a page to the importance of a site. The strength of the correlation may be based upon the depth of the page within the site. The importance system may iteratively calculate the importance of the pages using “conditional” transition probabilities. During each iteration, the importance system may recalculate the conditional transition probabilities based on the importance of sites that are derived from the recalculated importance of pages during the iteration.

Abstract: Systems and methods for processing user queries and identifying a set of documents relevant to the user query from a database using multi ranker search are described. In one implementation, the retrieved documents can be paired to form document pairs, or instance pairs, in a variety of combinations. Such instance pairs may have a rank order between them as they all have different ranks. A classifier, hyperplane, and a base ranker may be constructed for identifying the rank order relationships between the two instances in an instance pair. The base ranker may be generated for each rank pair. The systems use a divide and conquer strategy for learning to rank the instance pairs by employing multiple hyperplanes and aggregate the base rankers to form an ensemble of base rankers. Such an ensemble of base rankers can be used to rank the documents or instances.

Abstract: Method for determining a webpage importance, including receiving web browsing behavior data of one or more users; creating a model of the web browsing behavior data; calculating a stationary probability distribution of the model; and correlating the stationary probability distribution to the webpage importance.

Abstract: A “Ranking Optimizer,” provides a framework for directly optimizing conventional information retrieval (IR) measures for use in ranking, search, and recommendation type applications. In general, the Ranking Optimizer first reformats any conventional position based IR measure from a conventional “indexing by position” process to an “indexing by documents” process to create a newly formulated IR measure which contains a position function, and optionally, a truncation function. Both of these functions are non-continuous and non-differentiable. Therefore, the Ranking Optimizer approximates the position function by using a smooth function of ranking scores, and, if used, approximates the optional truncation function with a smooth function of positions of documents. Finally, the Ranking Optimizer optimizes the approximated functions to provide a highly accurate surrogate function for use as a surrogate IR measure.

Abstract: Importance ranking of web pages is performed by defining a graph-based regularization term based on document features, edge features, and a web graph of a plurality of web pages, and deriving a loss term based on human feedback data. The graph-based regularization term and the loss term are combined to obtain a global objective function. The global objective function is optimized to obtain parameters for the document features and edge features and to produce static rank scores for the plurality of web pages. Further, the plurality of web pages is ordered based on the static rank scores.

Abstract: Systems, methods, and devices for sorting and processing various types of graph data are described herein. Partitioning graph data into master data and associated slave data allows for sorting of the graph data by sorting the master data. In another embodiment, promoting a data bucket having a first data bucket size to a data bucket having a second data bucket size greater than the first data bucket size upon reaching a memory limit allows for the reduction of temporary files output by the data bucket.

Abstract: The page ranking technique described herein employs a Markov Skeleton Mirror Process (MSMP), which is a particular case of Markov Skeleton Processes, to model and calculate page importance scores. Given a web graph and its metadata, the technique builds an MSMP model on the web graph. It first estimates the stationary distribution of a EMC and views it as transition probability. It next computes the mean staying time using the metadata. Finally, it calculates the product of transition probability and mean staying time, which is actually the stationary distribution of MSMP. This is regarded as page importance.

Abstract: Many search engines attempt to understand and predict a user's search intent after the submission of search queries. Predicting search intent allows search engines to tailor search results to particular information needs of the user. Unfortunately, current techniques passively predict search intent after a query is submitted. Accordingly, one or more systems and/or techniques for actively predicting search intent from user browsing behavior data are disclosed herein. For example, search patterns of a user browsing a web page and shortly thereafter performing a query may be extracted from user browsing behavior. Queries within the search patterns may be ranked based upon a search trigger likelihood that content of the web page motivated the user to perform the query. In this way, query suggestions having a high search trigger likelihood and a diverse range of topics may be generated and/or presented to users of the web page.

Abstract: A method and system for rank aggregation of entities based on supervised learning is provided. A rank aggregation system provides an order-based aggregation of rankings of entities by learning weights within an optimization framework for combining the rankings of the entities using labeled training data and the ordering of the individual rankings. The rank aggregation system is provided with multiple rankings of entities. The rank aggregation system is also provided with training data that indicates the relative ranking of pairs of entities. The rank aggregation system then learns weights for each of the ranking sources by attempting to optimize the difference between the relative rankings of pairs of entities using the weights and the relative rankings of pairs of entities of the training data.

Abstract: A clustering system generates an original Laplacian matrix representing objects and their relationships. The clustering system initially applies an eigenvalue decomposition solver to the original Laplacian matrix for a number of iterations. The clustering system then identifies the elements of the resultant eigenvector that are stable. The clustering system then aggregates the elements of the original Laplacian matrix corresponding to the identified stable elements and forms a new Laplacian matrix that is a compressed form of the original Laplacian matrix. The clustering system repeats the applying of the eigenvalue decomposition solver and the generating of new compressed Laplacian matrices until the new Laplacian matrix is small enough so that a final solution can be generated in a reasonable amount of time.

Abstract: A calculate importance system calculates the global importance of a web page based on a “mean hitting time.” Hitting time of a target web page is a measure of the minimum number of transitions needed to land on the target web page. Mean hitting time of a target web page is an average number of such transitions for all possible starting web pages. The calculate importance system calculates a global importance score for a web page based on the reciprocal of a mean hitting time. A search engine may rank web pages of a search result based on a combination of relevance of the web pages to the search request and global importance of the web pages based on a global hitting time.