Abstract: Mechanisms are provided for transforming an original graph data set into a representative form having a smaller number of dimensions that the original graph data set. The mechanisms generate a graph transformation basis structure based on an input graph data structure. The mechanisms further transform an original graph data set based on an intersection of the graph transformation basis structure and the input graph data structure to thereby generate a transformed graph data set data structure. The transformed graph data set data structure has a reduced dimensionality from that of the input graph data structure but represents characteristics of the original graph data set. Moreover, the mechanisms perform an application specific operation on the transformed graph data set data structure to generate an output of a closest similarity record in the transformed graph data set to a target component.

Abstract: A method of labeling an unlabeled message of a social stream. The method including training a training model based on labeled messages, partitioning the training model into a plurality of class partitions, each comprising statistical information and a class label, computing a confidence for each of the class partitions based on information of an unlabeled message and the statistical information of a respective class partition, as executed by a processor in a computer system, and labeling the unlabeled message of the social stream according to respective confidences of the class partitions.

Abstract: Interoperability is enabled between participants in a network by determining values associated with a value metric defined for at least a portion of the network. Information flow is directed between two or more of the participants based at least in part on semantic models corresponding to the participants and on the values associated with the value metric. The semantic models may define interactions between the participants and define at least a portion of information produced or consumed by the participants. The determination of the values and the direction of the information flow may be performed multiple times in order to modify the one or more value metrics. The direction of information flow may allow participants to be deleted from the network, may allow participants to be added to the network, or may allow behavior of the participants to be modified.

Abstract: Improved privacy preservation techniques are disclosed for use in accordance with data mining. By way of example, a technique for preserving privacy of data records for use in a data mining application comprises the following steps/operations. Different privacy levels are assigned to the data records. Condensed groups are constructed from the data records based on the privacy levels, wherein summary statistics are maintained for each condensed group. Pseudo-data is generated from the summary statistics, wherein the pseudo-data is available for use in the data mining application.

Abstract: Mechanisms are provided for transforming an original graph data set into a representative form having a smaller number of dimensions that the original graph data set. The mechanisms generate a graph transformation basis structure based on an input graph data structure. The mechanisms further transform an original graph data set based on an intersection of the graph transformation basis structure and the input graph data structure to thereby generate a transformed graph data set data structure. The transformed graph data set data structure has a reduced dimensionality from that of the input graph data structure but represents characteristics of the original graph data set. Moreover, the mechanisms perform an application specific operation on the transformed graph data set data structure to generate an output of a closest similarity record in the transformed graph data set to a target component.

Abstract: Mechanisms are provided for transforming an original graph data set into a representative form having a smaller number of dimensions that the original graph data set. The mechanisms generate a graph transformation basis structure based on an input graph data structure. The mechanisms further transform an original graph data set based on an intersection of the graph transformation basis structure and the input graph data structure to thereby generate a transformed graph data set data structure. The transformed graph data set data structure has a reduced dimensionality from that of the input graph data structure but represents characteristics of the original graph data set. Moreover, the mechanisms perform an application specific operation on the transformed graph data set data structure to generate an output of a closest similarity record in the transformed graph data set to a target component.

Abstract: A method of labeling an unlabeled message of a social stream. The method including training a training model based on labeled messages, partitioning the training model into a plurality of class partitions, each comprising statistical information and a class label, computing a confidence for each of the class partitions based on information of an unlabeled message and the statistical information of a respective class partition, as executed by a processor in a computer system, and labeling the unlabeled message of the social stream according to respective confidences of the class partitions.

Abstract: A system that labels an unlabeled message of a social stream. The system including a memory device storing instructions to execute a training model, the training model being trained based on labeled messages, and partitioned into a plurality of class partitions, each of which comprise statistical information and a class label, and a Central Processing Unit (CPU) that computes a confidence for each of the class partitions based on information of an unlabeled message and the statistical information of a respective class partition, and that labels the unlabeled message according to respective confidences of the class partitions.

Abstract: A method (and system) for supervised network clustering includes receiving and reading node labels from a plurality of nodes on a network, as executed by a processor on a computer having access to the network, the network defined as a group of entities interconnected by links. The node labels are used to define densities associated with the nodes. Node components are extracted from the network, based on using thresholds on densities. Smaller components having a size below a user-defined threshold are merged.

Abstract: A method (and system) for supervised network clustering includes receiving and reading node labels from a plurality of nodes on a network, as executed by a processor on a computer having access to the network, the network defined as a group of entities interconnected by links. The node labels are used to define densities associated with the nodes. Node components are extracted from the network, based on using thresholds on densities. Smaller components having a size below a user-defined threshold are merged.

Abstract: Distributed privacy preserving data mining techniques are provided. A first entity of a plurality of entities in a distributed computing environment exchanges summary information with a second entity of the plurality of entities via a privacy-preserving data sharing protocol such that the privacy of the summary information is preserved, the summary information associated with an entity relating to data stored at the entity. The first entity may then mine data based on at least the summary information obtained from the second entity via the privacy-preserving data sharing protocol. The first entity may obtain, from the second entity via the privacy-preserving data sharing protocol, information relating to the number of transactions in which a particular itemset occurs and/or information relating to the number of transactions in which a particular rule is satisfied.

Abstract: Improved privacy preservation techniques are disclosed for use in accordance with data mining. By way of example, a technique for preserving privacy of data records for use in a data mining application comprises the following steps/operations. Different privacy levels are assigned to the data records. Condensed groups are constructed from the data records based on the privacy levels, wherein summary statistics are maintained for each condensed group. Pseudo-data is generated from the summary statistics, wherein the pseudo-data is available for use in the data mining application.

Abstract: Improved privacy preservation techniques are disclosed for use in accordance with data mining. By way of example, a technique for preserving privacy of data records for use in a data mining application comprises the following steps/operations. Different privacy levels are assigned to the data records. Condensed groups are constructed from the data records based on the privacy levels, wherein summary statistics are maintained for each condensed group. Pseudo-data is generated from the summary statistics, wherein the pseudo-data is available for use in the data mining application. Principles of the invention are capable of handling both static and dynamic data sets.

Abstract: Techniques for determining a shortest path in a disk-based network are provided. The techniques include creating a compressed representation of an underlying disk resident network graph, wherein creating a compressed representation of an underlying disk resident network graph comprises determining one or more dense regions in the disk resident graph and compacting the one or more dense regions into one or more compressed nodes, associating one or more node penalties with the one or more compressed nodes, wherein the one or more node penalties reflect a distance of a sub-path within a compressed node, and performing a query on the underlying disk resident network graph using the compressed representation and one or more node penalties to determine a shortest path in the disk-based network to reduce the number of accesses to a physical disk.

Abstract: Techniques for optimizing steady state flow of a network are provided. The techniques include determining a first set of two or more nodes in a network, computing a steady-state flow probability of the first set of two or more nodes, and iteratively interchanging nodes from a second set of two or more nodes into the first set of two or more nodes to determine an optimum total steady state flow of the network, wherein determining an optimum total steady-state flow of the network comprises iteratively interchanging nodes until no additional improvements in steady-state flow over the computed steady-state flow probability can be obtained.

Abstract: Techniques for identifying one or more communities in an information network are provided. The techniques include collecting one or more nodes and one or more edges from an information network, performing a random walk on the one or more nodes to produce a sequence of one or more nodes, creating a sequence database from one or more sequences produced via random walk, and mining the sequence database to determine one or more patterns in the network, wherein the one or more patterns identify one or more communities in the information network.

Abstract: A method for executing a query on a graph data stream. The graph stream comprises data representing edges that connect vertices of a graph. The method comprises constructing a plurality of synopsis data structures based on at least a subset of the graph data stream. Each vertex connected to an edge represented within the subset of the graph data stream is assigned to a synopsis data structure such that each synopsis data structure represents a corresponding section of the graph. The method further comprises mapping each received edge represented within the graph data stream onto the synopsis data structure which corresponds to the section of the graph which includes that edge, and using the plurality of synopsis data structures to execute the query on the graph data stream.

Abstract: An object and attributes that describe that object are identified. The attributes are grouped into attribute patterns, and classification classes are identified. For each identified class a sketch table containing a plurality of parallel hash tables is created. For the object to be classified, each attribute pattern is processed using the all of the hash functions for each sketch table, resulting in a plurality of values under each sketch table for a single attribute pattern. The lowest value is selected for each sketch table. The distribution of values across all sketch tables is evaluated for each attribute pattern, producing a discriminatory power for each attribute pattern. Attribute patterns having a discriminatory power above a given threshold are selected and added to the associated sketch table values. The sketch table with the largest overall sum is identified, and the associated class is assigned to the object belonging to the attribute patterns.

Abstract: Techniques for determining a shortest path in a disk-based network are provided. The techniques include creating a compressed representation of an underlying disk resident network graph, wherein creating a compressed representation of an underlying disk resident network graph comprises determining one or more dense regions in the disk resident graph and compacting the one or more dense regions into one or more compressed nodes, associating one or more node penalties with the one or more compressed nodes, wherein the one or more node penalties reflect a distance of a sub-path within a compressed node, and performing a query on the underlying disk resident network graph using the compressed representation and one or more node penalties to determine a shortest path in the disk-based network to reduce the number of accesses to a physical disk.

Abstract: Techniques for optimizing steady state flow of a network are provided. The techniques include determining a first set of two or more nodes in a network, computing a steady-state flow probability of the first set of two or more nodes, and iteratively interchanging nodes from a second set of two or more nodes into the first set of two or more nodes to determine an optimum total steady state flow of the network, wherein determining an optimum total steady-state flow of the network comprises iteratively interchanging nodes until no additional improvements in steady-state flow over the computed steady-state flow probability can be obtained.