International Business Machines Corporation

Chidanand V Apte

Kevin Patrick McAuliffe

Bonnie Kathryn Ray

Brenda Lynn Dietrich

Sholom M Weiss

Arun Hampapur

Andrew William Senior

Shubir Kapoor

Frederick J Damerau

Lea A Deleris

Brian F White

Eric W Cope

Bin Zhang

Philip Shilung Yu

Wen Jun Yin

Bishwaranjan Bhattacharjee

Charles Philippe Tresser

Sriram Kolinjivadi Padmanabhan

Jin Dong

Kenneth Alan Goldman

Ming Zhu

John Steven Langford

Ming Xie

Ta-Hsin Li

Naoki Abe

Matthias Gruetzner

Carl E Abrams

Matthew A Hilbert

Kathleen M Troidle

A governing modeling system maintains information associated with an analytic model. One or more policies may be defined that are associated with the analytic model and one or more instances of the analytic model. The system may monitor the analytic model and the one or more instances of the analytic model based on at least some of the information, the one or more policies associated with the analytic model and the one or more policies associated with the one or more instances of the analytic model.

Analytic model lifecycle maintenance and invalidation policies

Various embodiments for maintaining security and confidentiality of data and operations within a fraud detection system. Each of these embodiments utilizes a secure architecture in which: (1) access to data is limited to only approved or authorized entities; (2) confidential details in received data can be readily identified and concealed; and (3) confidential details that have become non-confidential can be identified and exposed.

Confidential fraud detection system and method

A method and system for quantifying risk by composing a heterogeneous risk model are provided. The method comprises providing risk input data associated with one or more risk nodes to a processor, running one or more risk models to output individual risk quantifications for each risk node, and aggregating the individual risk quantifications into a single output. The system comprises a processor operable to provide risk input data associated with one or more risk nodes to a processor, run one or more risk models to output individual risk quantifications for each risk node and aggregate the individual risk quantifications into a single output.

Method and system enabling dynamic composition of heterogenous risk models

A method and system for collaboratively managing risk is presented. In one embodiment the method comprises analyzing a node in a risk network when an output of said node violates a control limit, identifying a cause of said violation of said control limit, creating a mitigation action in response to said node violating said control limit, performing said mitigation action, said mitigation action having an influence on said output of said node, and measuring said output of said node to determine effectiveness of said mitigation action on said node, wherein a program using a processor unit executes one or more of said analyzing, identifying, creating, performing and measuring steps.

System and method for collaborative management of enterprise risk

Method, apparatus and system for location evaluation and site selection, capable of effectively configuring the site network and evaluating the facility location by scientifically modeling and incorporating human knowledge are provided. In one aspect, geographic and demographic data associated with a plurality of locations and human knowledge comprising partial rating knowledge and pair-wise preference knowledge are used in a regression algorithm to construct a location evaluation model. The regression algorithm is further refined using active learning that identifies a plurality of pairs of locations to improve precision of the regression algorithm.

Method and apparatus for location evaluation and site selection

Demographic and transaction data are presented in a report usable for marketing-related purposes and other applications using a camera apparatus to capture customer image data at a point-of-sale location and sends the image data to a customer image database where the image data is stored. Demographic data (such as gender, age group, height) are extracted from the customer image data and sent to the customer image database. Point-of-sale transaction data is collected and sent to a transaction database where the transaction data is stored. A computer processes customer image data, transaction data, and attributes relating to the customer image data and transaction data, to correlate extracted demographic data to transaction data. The correlated data is preferably sent to a target marketing database and used to generate a market research report.

Method and system for targeted marketing by leveraging video-based demographic insights

A procedure for clustering documents that operates in high dimensions, processes tens of thousands of documents and groups them into several thousand clusters or, by varying a single parameter, into a few dozen clusters. The procedure is specified in two parts: computing a similarity score representing the k most similar documents (typically the top ten) for each document in the collection, and grouping the documents into clusters using the similarly scores.

Lightweight document clustering

A lightweight document matcher employs minimal processing and storage. The lightweight document matcher matches new documents to those stored in a database. The matcher lists, in order, those stored documents that are most similar to the new document. The new documents are typically problem statements or queries, and the stored documents are potential solutions such as FAQs (Frequently Asked Questions). Given a set of documents, titles, and possibly keywords, an automatic back-end process constructs a global dictionary of unique keywords and local dictionaries of relevant words for each document. The application front-end uses this information to score the relevance of stored documents to new documents. The scoring algorithm uses the count of matched words as a base score, and then assigns bonuses to words that have high predictive value. It optionally assigns an extra bonus for a match of words in special sections, e.g., titles. The method uses minimal data structures and lightweight scoring algorithms to compute efficiently even in restricted environments, such as mobile or small desktop computers.

Light weight document matcher

A text categorization method automatically classifies electronic documents by developing a single pooled dictionary of words for a sample set of documents, and then generating a decision tree model, based on the pooled dictionary, for classifying new documents. Adaptive resampling techniques are applied to improve the accuracy of the decision tree model.

Method for improvement accuracy of decision tree based text categorization

Growing community of inventors

Yorktown Heights, NY, United States of America

Chidanand V Apte