Abstract: A method and system of providing training information for training a categorizer includes receiving a query relating to at least one category and identifying at least one case within a data set that matches the query. The method and system receives one of a first indication that the identified at least one case belongs to the category, and a second indication that the identified at least one case does not belong to the category. Training information is modified based on receiving one of the first indication and second indication.

Abstract: A technique for residual data identification can include receiving a plurality of data instances in a multi-class training data set that are d as belonging to recognized categories, receiving a plurality of data instances a first unlabeled data set, and receiving a plurality of data instances in a second unlabeled data set A technique for residual data identification can include labeling the plurality of data instances in the multi-class training data set as negative data instances. A technique for residual data identification can include labeling the plurality of data instances in the first unlabeled data set as positive data instances. A technique for residual data identification can include training a classifier with the labeled negative data instances and the labeled positive data instances. A technique for residual data identification can include applying the classifier to identify residual data instances in the second unlabeled data set.

Abstract: To detect duplicative hierarchically arranged sets of files in a storage system, a method includes generating, for hierarchically arranged plural sets of files, respective collections of values computed based on files in corresponding sets of files. For a further set of files that is an ancestor of at least one of the plural sets of files, a respective collection of values that is based on the collection of values computed for the at least one set is generated. Duplicative sets according to comparisons of the collections of values are identified.

Abstract: At least one case from a set of cases is identified in response to a query relating to at least one category. An indication is received regarding whether the at least one case belongs to the category. A categorizer is trained based on the received indication. A quantification measure associated with cases in the category is computed based on output from the categorizer.

Abstract: An apparatus and method convert user behavior data into rewards. The method and apparatus determine an undiluted reward for a user of a population of users. The method and apparatus divide the population of users into groups. The apparatus and method determine a dilution factor for each of the groups and assign and/or transmit a reward to the user based upon the undiluted reward and the dilution factor for the group to which that user belongs.

Abstract: A collection of labeled training cases is received, where each of the labeled training cases has at least one original feature and a label with respect to at least one class. Non-linear transformation of values of the original feature in the training cases is applied to produce transformed feature values that are more linearly related to the class than the original feature values. The non-linear transformation is based on computing probabilities of the training cases that are positive with respect to the at least one class. The transformed feature values are used to train a classifier.

Abstract: A categorizer produces a first measure regarding cases associated with an issue. Information regarding additional cases associated with the issue is received after one or more events have occurred with respect to the issue. Based on further output from the categorizer, a second measure is produced regarding the additional cases associated with the issue.

Abstract: To perform zone-based workload scheduling according to environmental conditions in a system having electronic devices, indicators of cooling efficiencies of the electronic devices in corresponding zones are aggregated to form aggregated indicators for respective zones, where the zones include respective subsets of electronic devices. Workload is assigned to the electronic devices according to the aggregated indicators.

Abstract: Methods, systems, and computer-readable and executable instructions are provided for assigning an advertisement. Assigning an advertisement can include constructing a graph of events between a number of visitors and a number of domains, iteratively calculating a set of domain scores for the advertisement using a spreading function on the graph, and assigning the advertisement to a domain based on a domain score of the domain, wherein the domain score is among the set of domain scores.

Abstract: For users in a set of users, corresponding measures of inferred interest in an item are determined. For a particular user, a particular measure of inferred interest in the item is determined. The particular measure of inferred interest and the measures of inferred interest corresponding to the users in the set of users are used to determine an estimate of the particular user's interest in the item.

Abstract: A classification count adjustment system for adjusting a count estimate of items in a dataset D classified into a class is disclosed. The system includes a count estimate produced by a classifier of the number of items in the dataset D classified into the class. The system further comprises one or more measures of behavior of the classifier, indicating the ability of the classifier to classify items into the class. The system further comprises a processor for computing an adjusted estimate based on the count estimate by the classifier and the one or more measures of behavior.

Abstract: A query-centric system and process for distributing reverse indices for a distributed content system. Relevance ranking techniques in organizing distributed system indices. Query-centric configuration subprocesses (1) analyze query data, partitioning terms for reverse index server(s) (RIS), (2) distribute each partitioned data set by generally localizing search terms for the RIS that have some query-centric correlation, and (3) generate and maintain a map for the partitioned reverse index system terms by mapping the terms for the reverse index to a plurality of different index server nodes. Indexing subprocess element builds distributed reverse indices from content host indices. Routines of the query execution use the map derived in the configuration to more efficiently return more relevant search results to the searcher.

Abstract: To perform zone-based workload scheduling according to environmental conditions in a system having electronic devices, indicators of cooling efficiencies of the electronic devices in corresponding zones are aggregated to form aggregated indicators for respective zones, where the zones include respective subsets of electronic devices. Workload is assigned to the electronic devices according to the aggregated indicators.

Abstract: A collection of labeled training cases is received, where each of the labeled training cases has at least one original feature and a label with respect to at least one class. Non-linear transformation of values of the original feature in the training cases is applied to produce transformed feature values that are more linearly related to the class than the original feature values. The non-linear transformation is based on computing probabilities of the training cases that are positive with respect to the at least one class. The transformed feature values are used to train a classifier.

Abstract: A query-centric system and process for distributing reverse indices for a distributed content system. Relevance ranking techniques in organizing distributed system indices. Query-centric configuration subprocesses (1) analyze query data, partitioning terms for reverse index server(s) (RIS), (2) distribute each partitioned data set by generally localizing search terms for the RIS that have some query-centric correlation, and (3) generate and maintain a map for the partitioned reverse index system terms by mapping the terms for the reverse index to a plurality of different index server nodes. Indexing subprocess element builds distributed reverse indices from content host indices. Routines of the query execution use the map derived in the configuration to more efficiently return more relevant search results to the searcher.

Abstract: The method discloses: scanning an original copy of a handwritten identifier; comparing the identifier to a set of prerecorded identifiers; adding the identifier to the set of prerecorded identifiers, if the identifier shares less than a predetermined similarity with any identifier within the set of prerecorded identifiers; linking the identifier to a set of information; scanning the original copy a second time; matching the second scan of the original copy with the first scan of the original copy; and retrieving the set of information in response to the matching. The system discloses: a computer system; a handwritten identifier; an imaging device for scanning an original copy of a handwritten identifier; an image processing and matching module for comparing the identifier to a set of prerecorded identifiers; a linking module for linking the identifier to a set of information.

Abstract: A system or method for minimally predictive feature identification is disclosed. For information management, an information collection including a set of features is received. A set of prediction values indicating a degree to which a first feature within the set of features predicts other features in the set is generated. The first feature as a minimally predictive feature is identified if each of the prediction values is within a predetermined range of threshold values.

Abstract: A query-centric system and process for distributing reverse indices for a distributed content system. Relevance ranking techniques in organizing distributed system indices. Query-centric configuration subprocesses (1) analyze query data, partitioning terms for reverse index server(s) (RIS), (2) distribute each partitioned data set by generally localizing search terms for the RIS that have some query-centric correlation, and (3) generate and maintain a map for the partitioned reverse index system terms by mapping the terms for the reverse index to a plurality of different index server nodes. Indexing subprocess element builds distributed reverse indices from content host indices. Routines of the query execution use the map derived in the configuration to more efficiently return more relevant search results to the searcher.

Abstract: A method, apparatus, and system identity files with similar content. One embodiment is a method that divides files into plural segments. The method computes a hash value and a size for each of the plural segments of the files. In order to identify which files have similar content, the method adds together segments common between files. File similarity information of files with similar content is output.

Abstract: A file is divided into plural segments, and for each of at least some of the plural segments, plural windows covering different portions of the corresponding segment are defined. According to a predefined criterion, one of the windows associated with the corresponding segment is selected. A representative hash based on the portion in the selected window is calculated for representing the corresponding segment.