International Business Machines Corporation

At&t Intellectual Property I, LP

Hanghang Tong

Ching-Yung Lin

Jimeng Sun

Leman Akoglu

U Kang

Fei Wang

Hani Talal Jamjoom

Marcel Catalin Rosu

Huiming Qu

Ravindranath Konuru

David M Lubensky

Yinglong Xia

Zhen Wen

Yehuda Koren

Christos Faloutsos

Ravi B Konuru

Jingrui He

Spyridon Papadimitriou

Danai Koutra

A method for determining a similarity between a plurality of graphs includes inferring a low-rank representation of a first graph, inferring a low-rank representation of a second graph, wherein the low-rank representations of the first and second graphs are stored in memory, estimating a left interaction between the first and second graphs, estimating a middle interaction between the first and second graphs, estimating a right interaction between the first and second graphs, wherein the estimations are based on the low-rank representations of the first and second graphs stored in memory, and aggregating the left interaction, the middle interaction and the right interaction into a kernel, wherein the kernel is indicative of the similarity between the first and second graphs.

Determining a similarity between graphs

Techniques are provided for anomaly, association and clustering detection. At least one code table is built for each attribute in a set of data. A first code table corresponding to a first attribute and a second code table corresponding to a second attribute are selected. The first code table and the second code table are merged into a merged code table, and a determination is made to accept or reject the merged code table. An anomaly is detected when a total compression cost for a data point is greater than a threshold compression cost inferred from one or more code tables. An association in a data table is detected by merging attribute groups, splitting data groups, and assigning data points to data groups. A cluster is inferred from a matrix of data and code words for each of the one or more code tables.

Anomaly, association and clustering detection

A method for determining a correspondence between a first node set of a first graph and a second node set of a second graph includes building a feature representation for each of the first graph and the second graph, and inferring the correspondence between the first node set and the second node set based on the feature representations.

Determining soft graph correspondence between node sets based on feature representations

Methods and systems for inferring performance include a chain graph that has multiple nodes, one or more of which has a known value at a first timestamp. A value associated with a first node at the first timestamp influences a value associated with a second node connected to the first node by an edge at the first timestamp. Each node has an associated chain that represents values of the node at different timestamps. Known values of each node in the graph are propagated within each timestamp to all neighboring nodes. Known values of each node in the graph are propagated across different timestamps.

Characterizing success pathways in networked graphs

A method of building a soft linkage between a plurality of graphs includes initializing a correspondence between type-1 and type-2 objects in the plurality of graphs, and reducing a cost function by alternately updating the type-1 correspondence and updating the type-2 correspondence.

Graph matching

Techniques are provided for dynamic anomaly, association and clustering detection. At least one code table is built for each attribute in a set of data containing one or more attributes. One or more clusters associated with one or more of the code tables are established. One or more new data points are received. A determination is made if a given one of the new data points is an anomaly. At least one of the one or more code tables is updated responsive to the determination. When a compression cost of a given one of the new data points is greater than a threshold compression cost for each of the one or more clusters, the given one of the new data points is an anomaly.

Dynamic anomaly, association and clustering detection

A method, system and computer program product for finding a diversified ranking list for a given query. In one embodiment, a multitude of date items responsive to the query are identified, a marginal score is established for each data item; and a set, or ranking list, of the data items is formed based on these scores. This ranking list is formed by forming an initial set, and one or more data items are added to the ranking list based on the marginal scores of the data items. In one embodiment, each of the data items has a measured relevance and a measured diversity value, and the marginal scores for the data items are based on the measured relevance and the measured diversity values of the data items.

Finding a top-K diversified ranking list on graphs

Computer-implemented methods, systems, and articles of manufacture for determining the importance of a data item. A method includes: (a) receiving a node graph; (b) approximating a number of neighbor nodes of a node; and (c) calculating a average shortest path length of the node to the remaining nodes using the approximation step, where this calculation demonstrates the importance of a data item represented by the node. Another method includes: (a) receiving a node graph; (b) building a decomposed line graph of the node graph; (c) calculating stationary probabilities of incident edges of a node graph node in the decomposed line graph, and (d) calculating a summation of the stationary probabilities of the incident edges associated with the node, where the summation demonstrates the importance of a data item represented by the node. Both methods have at least one step carried out using a computer device.

Determining the importance of data items and their characteristics using centrality measures

Access is obtained to a first nonnegative factor matrix and a second nonnegative factor matrix obtained by factorizing a nonnegative asymmetric matrix which represents a set of data which tracks time-stamped activities of a plurality of entities. The first nonnegative factor matrix is representative of initial role membership of the entities, and the second nonnegative factor matrix is representative of initial role activity descriptions. At a given one of the time stamps, while holding a change in the first nonnegative factor matrix constant, a change in the second nonnegative factor matrix is updated to reflect time variance of the set of data at the given one of the time stamps, without accessing actual data values at previous ones of the time stamps. At the given one of the time stamps, while holding a change in the second nonnegative factor matrix constant, a change in the first nonnegative factor matrix is updated, to reflect the time variance of the set of data at the given one of the time stamps, without accessing the actual data values at the previous ones of the time stamps. The role membership of the entities and the role activity descriptions, at the given one of the time stamps, are updated based on the updating steps. A suitable technique for nonnegative symmetric matrices is also provided.

Privacy-aware on-line user role tracking

An embodiment of the invention pertains to a weighted directed graph comprising multiple nodes and edges that each extends between two nodes. The embodiment includes processing edges to generate a forward and reverse edge corresponding to each edge. Forward and reverse edges are processed to generate indirect edges, each comprising two edge components, and extending between two nodes. One node associated with each forward edge, each reverse edge, and each indirect edge is selected to be the key node of its associated edge. All forward, reverse and indirect edges having a particular node as their respective key nodes are placed into a group. All edges of the group are then selectively processed to provide information pertaining to an egonet of the graph that has the particular node as its egonode.

Efficient egonet computation in a weighted directed graph

A method, system and computer program product for managing and querying a graph. The method includes the steps of: receiving a graph; partitioning the graph into homogeneous blocks; compressing the homogeneous blocks; and storing the compressed homogeneous blocks in files where at least one of the steps is carried out using a computer device.

Method and system for managing and querying large graphs

A method of detecting anomalies from a bipartite graph includes analyzing the graph to determine a row-cluster membership, a column-cluster membership and a non-negative residual matrix, and in a processor, detecting the anomalies from the non-negative residual matrix.

Growing community of inventors

Ossining, NY, United States of America

Hanghang Tong