International Business Machines Corporation

Jun Li Wang

Kush Raj Varshney

Fei Wang

Aleksandra Mojsilovic

Lav Raj Varshney

Jinfeng Yi

Hua Wang

Shuo Li

Xin Peng Liu

Sheng Yan Sun

Mark A Takacs

Rebecca Yung Gorrell

Xu Sun

Kenneth J Travis, Jr

Bruce Anthony Copeland

He Huang

Fei C Wang

Jianying Hu

Li Ni Zhang

Wen Sun

Amit Dhurandhar

Guan Zhang

Karthikeyan Natesan Ramamurthy

Sonia Rodriguez

Jon Charles Lorenz

Thomas Gabriel Tampone

Lovell B Wiggins

Buyue Qian

Hongling Zhang

Li Mei Zhang

Laurence J Nummy

Pengtao Zhang

Yu Zhuang

Xiang Wang

Dongping Fang

Joshua D Sieber

Phouvieng Beyer

Kanwar Pal Singh Sidhu

Jolaine Savoie

John H Bauer

Jason A Brazzillo

Magnus Eriksson

Zhulin Tan

Weitong Dong

Peter O Ulanmo

Donald W Scheider

The present invention relates to compounds of (II) and an acceptable salt or hydrate thereof method of making same.

Intermediate compounds and methods

Systems and methods for indexing data include formulating an objective function to index a dataset, a portion of the dataset including supervision information. A data property component of the objective function is determined, which utilizes a property of the dataset to group data of the dataset. A supervised component of the objective function is determined, which utilizes the supervision information to group data of the dataset. The objective function is optimized using a processor based upon the data property component and the supervised component to partition a node into a plurality of child nodes.

Indexing of large scale patient set

Implementations of the present disclosure relate to methods, systems, and computer program products for security management. In one implementation, a computer-implemented method is disclosed. In the method, a message sequence associated with a user may be obtained from a list of historical messages that are logged in a data system. A candidate operation may be determined based on the obtained message sequence and an association model, where the association model indicates associations between message sequences and operations that are supported in the data system. A security level of the candidate operation may be evaluated based on at least one historical operation that has been performed by the user in the data system. In other implementations, a computer-implemented system and a computer program product for security management are disclosed.

Security management for data systems

A system and method for patient stratification include determining a first set of patient groups from patients in a patient similarity graph based on a similarity structure of the patient similarity graph. A second set of patient groups is identified based on expert domain knowledge associated with the patients. Patients in the first set and the second set are aligned using a processor to stratify patients.

Patient risk stratification by combining knowledge-driven and data-driven insights

Embodiments of the present disclosure relate to a computer-implemented method for executing one or more operations based on a command. According to the method, a command is received. A command name of the command and one or more parameters of the command are extracted. A command description file corresponding to the command is retrieved based on the command name. The command description file describes function logic of the command. The retrieved command description file is parsed, and information for executing one or more operations included in the command description file is acquired. The one or more operations are executed based on the acquired information and the one or more parameters.

Command tool development using a description file

Methods and systems for generating prediction data are described. In an example, a processor may retrieve preferential data from a memory. The preferential data may include a set of preferences that corresponds to a first subset of objects among a set of objects, and may exclude preferences associated with a second subset of objects among the set of objects. Each preference may indicate a preferred object between a respective pair of objects among the first subset of objects. The processor may determine first predicted ratings corresponding to the first subset of objects based on the preferential data. The processor may determine second predicted ratings corresponding to the second subset of objects based on the preferential data. The processor may generate prediction data by populating entries of the prediction data with the first and second predicted ratings, where the prediction data may include predicted ratings of the set of objects.

Recommendation prediction based on preference elicitation

A method includes performing contextual association of entities using multi-source data. For each context the method performs co-clustering to identify distinct expert-skill associations; constructing single-entity unipartite graph representations and performing a random walk within each single-entity unipartite graph; for each single-entity unipartite graph, obtaining steady state distributions using the random walks to obtain clusters of experts and skills; performing a weighted two-way random walk across entity graphs (graph edges), giving preference to traversal within members of the same co-cluster; and performing link prediction for each context by dynamically adding edges, and obtaining overall skills predictions, analyses and inferences by merging the contexts and weighting the links of each context. The method can also use the context-specific weights obtained from the co-association information in a matrix completion procedure, and finally merge the context-specific outputs to obtain overall skills predictions, analyses and inferences. A computer program product and a system are also disclosed for performing the method.

Method, system and computer program product for automating expertise management using social and enterprise data

A system, method and computer program product for generating a classification model using original data that is sensitive or private to a data owner. The method includes: receiving, from one or more entities, a masked data set having masked data corresponding to the original sensitive data, and further including a masked feature label set for use in classifying the masked data contents; forming a shared data collection of the masked data and the masked feature label sets received; and training, by a second entity, a classification model from the shared masked data and feature label sets, wherein the classification model learned from the shared masked data and feature label sets is the same as a classification model learned from the original sensitive data. The sensitive features and labels cannot be reliably recovered even when both the masked data and the learning algorithm are known.

Privacy and modeling preserved data sharing

A method for generating a classification model using original data that is sensitive or private to a data owner. The method includes: receiving, from one or more entities, a masked data set having masked data corresponding to the original sensitive data, and further including a masked feature label set for use in classifying the masked data contents; forming a shared data collection of the masked data and the masked feature label sets received; and training, by a second entity, a classification model from the shared masked data and feature label sets, wherein the classification model learned from the shared masked data and feature label sets is the same as a classification model learned from the original sensitive data. The sensitive features and labels cannot be reliably recovered even when both the masked data and the learning algorithm are known.

Determining a data security risk level of a virtual database is provided. An object catalog corresponding to a real database is imported into the virtual database. Objects in the object catalog are organized by levels. It is determined whether one or more data security policy definitions corresponding to a set of objects referenced by test query message traffic performed an action in response to determining that one or more test query messages in the test query message traffic run on the virtual database did not satisfy respective parameters of the one or more data security policy definitions. In response to determining that one or more of the data security policy definitions corresponding to the set of objects referenced by the test query message traffic did not perform the action, a test failure result is returned. A data security risk level for the virtual database is determined based on the result.

Generating a virtual database to test data security of a real database

A method for designing a new product includes automatically identifying a plurality of components of stored products, wherein the plurality of components is associated with an existing component included in a design of the new product, automatically selecting a new component from among the plurality of components, wherein a relationship between the existing component and the new component optimizes a design objective for the new product, automatically adding the new component to the design for the new product, and iteratively repeating the identifying, the selecting, and the adding until a stopping criterion is met, wherein the design for the new product is complete when the stopping criterion is met.

Association-based product design

A method of performing a search of heterogeneous data based on an input query includes: generating an index including at least two hash tables, where each hash table corresponds to a different data domain of the heterogeneous data and includes hash code sets, where at least one of the hash code sets is mapped to a hash code set of another one of the tables. The method further includes performing a hash on the input query to generate a hash code, by referring to the index, determining a first hash code set that the generated hash code belongs to, and determining a second hash code set that the determined first hash code set is mapped to, and providing at least one result based on the determined second hash code set.

Indexing and searching heterogenous data entities

Growing community of inventors

Bellevue, WA, United States of America

Jun Li Wang