Sholom M Weiss

6523020, Feb 18, 2003

Mar 22, 2000

09/533323

Sholom Weiss - Highland Park NJ

International Business Machines Corporation - Armonk NY

G06N 502

706 47, 707 5

A lightweight rule induction method is described that generates compact Disjunctive Normal Form (DNF) rules. Each class may have an equal number of unweighted rules. A new example is classified by applying all rules and assigning the example to the class with the most satisfied rules. The induction method attempts to minimize the training error with no pruning. An overall design is specified by setting limits on the size and number of rules. During training, cases are adaptively weighted using a simple cumulative error method. The induction method is nearly linear in time relative to an increase in the number of induced rules or the number of cases. Experimental results on large benchmark datasets demonstrate that predictive performance can rival the best reported results in the literature.

6654739, Nov 25, 2003

Jan 31, 2000

09/494744

Chidanand Apte - Chappaqua NY
Sholom M. Weiss - Highland Park NJ
Brian F. White - Yorktown Heights NY

International Business Machines Corporation - Armonk NY

G06F 1730

707 5, 707200, 707205

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.

7539585, May 26, 2009

Jun 14, 2007

11/763052

Robert J. Baseman - Brewster NY, US
Fateh A. Tipu - Wappingers Falls NY, US
Sholom M. Weiss - New York NY, US

International Business Machines Corporation - Armonk NY

G01N 37/00

702 81, 702179, 700 9, 700 28, 700 32, 700108

A fabrication history of a group of wafers is provided, having a record for each wafer of the manufacturing events that did or did not occur in its fabrication, and having the measured value of a given target. A binary decision rule is formed based on the fabrication history, the rule being that if a wafer has a particular pattern of manufacturing events in its fabrication history then the statistic of the given fabrication target for that wafer is a first value; otherwise, the statistic is a second value having at least a given distance from the first value. The pattern of manufacturing events in the binary decision rule is identified in the generation of the binary decision rule. The identified pattern is significant with respect to the given target.

7650251, Jan 19, 2010

Dec 12, 2008

12/333830

Robert J. Baseman - Brewster NY, US
Fateh A. Tipu - Wappingers Falls NY, US
Sholom M. Weiss - New York NY, US

International Business Machines Corporation - Armonk NY

G01N 37/00

702 81, 702179, 700 9, 700 28, 700 32, 700108

A fabrication history a group of wafers is provided, having a record for each wafer of the manufacturing events that did or did not occur in its fabrication, and having the measured value of a given target. A binary decision rule is formed based on the fabrication history, the rule being that if a wafer has a particular pattern of manufacturing events in its fabrication history then the statistic of the given fabrication target for that wafer is a first value; otherwise, the statistic is a second value having at least a given distance from the first value. The pattern of manufacturing events in the binary decision rule is identified in the generation of the binary decision rule. The identified pattern is significant with respect to the given target.

20030033436, Feb 13, 2003

May 14, 2001

09/853620

Sholom Weiss - New York NY, US

G06F009/00
G06F009/54
G06F015/163

709/310000

A pattern recognition method induces ensembles of decision rules from data regression problems. Instead of direct prediction of a continuous output variable, the method discretizes the variable by k-means clustering and solves the resultant classification problem. Predictions on new examples are made by averaging the mean values of classes with votes that are close in number to the most likely class.

62860003, Sep 4, 2001

Dec 1, 1998

9/203673

Chidanand Apte - Chappaqua NY
Frederick J. Damerau - North Salem NY
Sholom M. Weiss - Highland Park NJ
Brian F. White - Yorktown Heights NY

International Business Machines Corporation - Armonk NY

G06F 1730

707 5

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.

57242631, Mar 3, 1998

May 30, 1997

8/866047

Sasisekharan Raguram - Somerset NJ
V. Seshadri - Lincroft NJ
Sholom M. Weiss - Highland Park NJ

AT&T Corp - Middletown NJ

G06F 1518

364552

A facility is provided for enhancing an operations support system so that, based on data generated as a result of an event occurring in an associated telecommunications network, the operations support system can predict the likelihood of the event occurring again in the network.

62531697, Jun 26, 2001

May 28, 1998

9/084985

Chidanand Apte - Chappaqua NY
Frederick J. Damerau - North Salem NY
Sholom M. Weiss - Highland Park NJ

International Business Machines Corporation - Armonk NY

G06E 1727
G06E 2100

704 9

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.