Primal Fusion Inc.

Tamr, Inc.

International Business Machines Corporation

Other

Ihab Francis Ilyas

Peter Joseph Sweeney

George Beskales

Mohamed A Soliman

Daniel Meir Bruckner

Nikolaus Bates-Haus

Nadiya Yampolska

Michael Ralph Stonebraker

Mathew Whitney Wilson

Alexander Richter Pagan

Liam Cleary

Gideon Goldin

Sharon Roth

John Kraemer

Anne Jude Hunt

Paul Roome

Jean-Paul Dupuis

Calisto Paul Zuzarte

Alexander David Black

Amr El-Helw

Naim Khan

Andrew Russell

Wu Zhou

Wayne Oldford

Volker Gerhard Markl

George Anwar Dany Beskales

Jason Liu

Wing Yan Lau

Tianyu Zhu

Kanak Kshetri

Vladimir Gluzman Peregrine

Stan Zdonik

Andrew H Palmer

Aizana Turmukhametova

A system and method of use resolves the frustration of repeated manual work during schema mapping. The system utilizes a transformation graph—a collection of nodes (unified attributes) and edges (transformations) in which source attributes are mapped and transformed. The system further leverages existing mappings and transformations for the purpose of suggesting to a user the optimal paths (i.e., the lowest cost paths) for mapping new sources, which is particularly useful when new sources share similarity with previously mapped sources and require the same transformations. As such, the system also promotes an evolving schema by allowing users to select which unified attributes they want to include in a target schema at any time. The system addresses the technical challenge of finding optimal transformation paths and how to present these to the user for evaluation.

Generating and reusing transformations for evolving schema mapping

Techniques for presenting content to users. The techniques include: obtaining user context information including a first keyword; identifying, based on the first keyword, a first attribute and a second attribute among the plurality of attributes, the first attribute being a characteristic of the first keyword and the second attribute being another characteristic of the first keyword; obtaining, based on the user context information, at least one second-order user preference among attributes in the plurality of attributes including a preference between the first attribute and the second attribute; identifying a set of content items among the plurality of content items based on the first attribute and the second attribute; determining a ranking of content items in the set of content items based on the at least one second-order user preference; and presenting content items to the user in accordance with the ranking.

Techniques for presenting content to a user based on the user's preferences

Systems and methods are provided for generating training data for a machine-learning classifier. A knowledge representation synthesized based on an object of interest is used to assign labels to content items. The labeled content items can be used as training data for training a machine learning classifier. The labeled content items can also be used as validation data for the classifier.

System and method for generating training data for machine learning classifier

Techniques for use in identifying one of more concepts in a knowledge representation (KR). The techniques include obtaining user context information associated with a user, wherein the user context information comprises a plurality of words; Also included are semantic disambiguation techniques comprising obtaining user context information associated with a user, wherein the user context information comprises a first portion and a second portion different from the first portion; and disambiguating between a first and second concept in a knowledge representation (KR) associated with a first meaning of the first portion. Semantic disambiguation techniques further include obtaining user context information associated with a user, wherein the user context information comprises a first portion and a second portion different from the first portion; and disambiguating between a first concept and second concept in a knowledge representation (KR) using a measures of dominance and semantic coherence. Additionally, techniques are disclosed for calculating a measure of semantic coherence based on a graph of a knowledge representation (KR) and, an overlap of semantic context of a first concept and a second concept in the KR.

Methods and apparatus for identifying concepts corresponding to input information

Techniques for customizing knowledge representation systems including identifying, based on a plurality of concepts in a knowledge representation (KR), a group of one or more concepts relevant to user context information, and providing the identified group of one more concepts to a user. The KR may include a combination of modules. The modules may include a kernel and a customized module customized for the user. The kernel may accessible via a second KR.

Methods and devices for customizing knowledge representation systems

Record clustering is performed for a collection of records using training rules, training-rule labels, training data created from a sample of pairs of records, a pair-wise classifier, and a clustering algorithm. Record clustering is also performed for a collection of records using prediction rules, prediction-rule labels, a pair-wise classifier, and a clustering algorithm.

Methods and computer program products for clustering records using imperfect rules

Given a number of records and a number of target classes to which these records belong to, a (weakly) supervised machine learning classification method leverages known possibly dirty classification rules, efficiently and accurately learns a classification model from training data, and applies the learned model to the data records to predict their classes.

Computer-implemented method for performing hierarchical classification

An end-to-end data curation system and the various methods used in linking, matching, and cleaning large-scale data sources. The goal of this system is to provide scalable and efficient record deduplication. The system uses a crowd of experts to train the system. The system operator can optionally provide a set of hints to reduce the number of questions send to the experts. The system solves the problem of schema mapping and record deduplication a holistic way by unifying these problems into a unified linkage problem.

Method and system for large scale data curation

Fast record deduplication is accomplished by providing as an input, data records having multiple attributes, and local similarity functions of individual attributes with local similarity thresholds. Bin IDs are then generated based on the local similarity functions and the local similarity thresholds. The Bin IDs are unique identifiers of a respective bin of records, and the bin of records is a set of records that are possibly pairwise similar. Local candidate pairs are identified based on data records that share Bin IDs. The local candidate pairs are aggregated to produce a set of global candidate pairs. The set of global candidate pairs are filtered by deciding whether a pair of data records represents a duplicate.

Scalable binning for big data deduplication

Growing community of inventors

Waterloo, Canada

Ihab Francis Ilyas