Abstract: A data processing method includes receiving data from a plurality of sources, generating a knowledge graph, and identifying potentially ambiguous particular nodes. For each of the particular nodes a semantic neighborhood is constructed including the particular node and a plurality of nodes within a particular degree of separation from the particular node, the particular node is omitted from the semantic neighborhood, a plurality of groups of connected nodes are determined in the semantic neighborhood, a plurality of instances of the particular node are defined, a plurality of substitute nodes are generated based on the plurality of instances, and each of the plurality of groups of connected nodes are connected to one of the plurality of substitute nodes to update the knowledge graph. A query is received via a network. Query results are retrieved out of the updated knowledge graph, and the query results are rendered accessible.

Abstract: A method includes accessing a database including a table including a first column including a first plurality of values, a second column including a second plurality of values, and a plurality of rows. A knowledge graph is constructed including a first plurality of nodes based on the first plurality of values, a second plurality of nodes based on the second plurality of values, and a plurality of node connections. A missing value is detected in the second column. A first particular value of the first plurality of values is detected in the first column. A first particular node corresponding to the first particular value is detected. The first particular node is determined to be connected to a second particular node corresponding to a second particular value of the second plurality of values, and the missing value is filled with the second particular value.

Abstract: A data processing method includes receiving data from a plurality of sources, generating a knowledge graph, and identifying potentially ambiguous particular nodes. For each of the particular nodes a semantic neighborhood is constructed including the particular node and a plurality of nodes within a particular degree of separation from the particular node, the particular node is omitted from the semantic neighborhood, a plurality of groups of connected nodes are determined in the semantic neighborhood, a plurality of instances of the particular node are defined, a plurality of substitute nodes are generated based on the plurality of instances, and each of the plurality of groups of connected nodes are connected to one of the plurality of substitute nodes to update the knowledge graph. A query is received via a network. Query results are retrieved out of the updated knowledge graph, and the query results are rendered accessible.

Abstract: A method for resolving entities in a knowledge graph including determining node sets in the knowledge graph, determining each of the node sets includes determining a first node, determining a second node in a semantic neighborhood of the first node, and determining a third node in the semantic neighborhood of the first node. For each node set, the second node and the third node are compared, and it is determined that the second node and the third node are a similar node pair. For each similar node pair, the first nodes of the node sets are aggregated, and a quantity of overlapping of a semantic neighborhood of the second node and a semantic neighborhood of the third node is determined, and for each similar node pair, the second and third nodes are resolved as a single entity.

Abstract: A method of machine learning for generating a predictive model of a response characteristic based on historical data elements using a processor may include receiving historical data elements and historical values for the response characteristic related to uses of the historical data elements in web pages. A plurality of key-value pairs may be generated defining values of a plurality of predefined features representing properties of the historical data elements. Each of a plurality of n features may be represented by an axis in an n-dimensional space are extracted from the historical data elements. The extracted plurality of key-value pairs for each historical data element may be projected onto the n-dimensional space. The plurality of vectors may be input into a model generator to generate a predictive model predicting a value of the response characteristic for a new data element.

Abstract: A method of classifying data is disclosed. Text data items are received. A set of classes into which the text data items are to be classified is received. A phrase-based classifier to classify the text data items into the set of classes is selected. The phrase-based classifier is applied to classify the text data items into the classes. Here, the applying includes creating a controlled vocabulary pertaining to classifying the text data items into the set of classes, building phrases based on the text data items and the controlled vocabulary, and classifying the text data items into the set of classes based on the phrases.

Abstract: A method of classifying data is disclosed. Text data items are received. A set of classes into which the text data items are to be classified is received. A phrase-based classifier to classify the text data items into the set of classes is selected. The phrase-based classifier is applied to classify the text data items into the classes. Here, the applying includes creating a controlled vocabulary pertaining to classifying the text data items into the set of classes, building phrases based on the text data items and the controlled vocabulary, and classifying the text data items into the set of classes based on the phrases.

Abstract: A method of classifying data is disclosed. Text data items are received. A set of classes into which the text data items are to be classified is received. A phrase-based classifier to classify the text data items into the set of classes is selected. The phrase-based classifier is applied to classify the text data items into the classes. Here, the applying includes creating a controlled vocabulary pertaining to classifying the text data items into the set of classes, building phrases based on the text data items and the controlled vocabulary, and classifying the text data items into the set of classes based on the phrases.

Abstract: A method of classifying data is disclosed. Text data items are received. A set of classes into which the text data items are to be classified is received. A phrase-based classifier to classify the text data items into the set of classes is selected. The phrase-based classifier is applied to classify the text data items into the classes. Here, the applying includes creating a controlled vocabulary pertaining to classifying the text data items into the set of classes, building phrases based on the text data items and the controlled vocabulary, and classifying the text data items into the set of classes based on the phrases.

Abstract: A technique for determining the impact of multiple users' actions is described. According to this technique, information specifying multiple users' actions is aggregated. This information includes linkages that indicate the initiators of the multiple users' actions. Then, the impacts of the multiple users' actions are determined based on the aggregated information. In particular, the impact of a given user's action includes a direct component performed by the given user and an indirect component performed by other users in a social graph in response to the given user's action, where the social graph includes relationships between the other users and the given user. For example, the multiple users' actions and the social graph may be associated with a website, and the determined impacts may allow traffic and/or revenue of the website to be increased.

Abstract: Improving data clustering stability. A computer accesses a first plurality of cluster groups comprising data. The computer then applies a clustering method to the first plurality of cluster groups while adjusting said first plurality of cluster groups to be in higher agreement between themselves, thereby generating a second plurality of cluster groups that is in higher agreement between themselves than the first plurality of cluster groups. The second plurality of cluster groups corresponds to the first plurality of cluster groups.

Abstract: A method of classifying data is disclosed. Text data items are received. A set of classes into which the text data items are to be classified is received. A phrase-based classifier to classify the text data items into the set of classes is selected. The phrase-based classifier is applied to classify the text data items into the classes. Here, the applying includes creating a controlled vocabulary pertaining to classifying the text data items into the set of classes, building phrases based on the text data items and the controlled vocabulary, and classifying the text data items into the set of classes based on the phrases.

Abstract: A method for data clustering may comprise entering data into a computer network comprising a master processor, an array of slave processors, and two cluster seats associated with each slave processor; executing a master process comprising dividing the data into clusters, sending the clusters to the cluster seats, initializing an optimization cycle, and computing an objective function. The optimization cycle includes the parallel execution by the slave processors of a slave process, which includes exchanging data between paired clusters so as to increase the objective function based on two modalities, and then resorting the cluster pairs for a subsequent iteration of the process.

Abstract: One embodiment is a method that builds a website profile from keywords appearing at the website and builds a user profile from a subset of the keywords that appear in documents accessed by the user. A web page is personalized based on the user profile.

Abstract: A method for data clustering may comprise entering data into a computer network comprising a master processor, an array of slave processors, and two cluster seats associated with each slave processor; executing a master process comprising dividing the data into clusters, sending the clusters to the cluster seats, initializing an optimization cycle, and computing an objective function. The optimization cycle includes the parallel execution by the slave processors of a slave process, which includes exchanging data between paired clusters so as to increase the objective function based on two modalities, and then resorting the cluster pairs for a subsequent iteration of the process.