Rakesh AgrawalSan Jose, CA

6370526, Apr 9, 2002

May 18, 1999

09/314479

Rakesh Agrawal - San Jose CA
Andreas Arning - Rottenburg, DE
Roland Seiffert - Herrenberg, DE
Ramakrishnan Srikant - San Jose CA

International Business Machines Corporation - Armonk NY

G06F 1730

707 5

A method and system for presenting a group of objects in a ranking order. Objects are ranked according to user preferences by first observing the access order of a related group of objects in relation to a predetermined access hypothesis. A user preference model is then adapted to correspond to any deviations between the access order and the access hypothesis for the related group of objects. Next, object preferences are calculated for each of the objects to be ranked according to the preference model. The group of objects is then presented to the user in an order corresponding to the calculated object preferences. The preference model is adaptively updated, unbeknownst to the user, in the normal course of accessing the presented objects.

6546389, Apr 8, 2003

Jan 19, 2000

09/487643

Rakesh Agrawal - San Jose CA
Ramakrishnan Srikant - San Jose CA

International Business Machines Corporation - Armonk NY

G06F 1730

707 6, 707101, 707102

A system and method for mining data while preserving a users privacy includes perturbing user-related information at the users computer and sending the perturbed data to a Web site. At the Web site, perturbed data from many users is aggregated, and from the distribution of the perturbed data, the distribution of the original data is reconstructed, although individual records cannot be reconstructed. Based on the reconstructed distribution, a decision tree classification model or a Naive Bayes classification model is developed, with the model then being provided back to the users, who can use the model on their individual data to generate classifications that are then sent back to the Web site such that the Web site can display a page appropriately configured for the users classification. Or, the classification model need not be provided to users, but the Web site can use the model to, e. g. , send search results and a ranking model to a user, with the ranking model being used at the user computer to rank the search results based on the users individual classification data.

6592627, Jul 15, 2003

Jun 10, 1999

09/329684

Rakesh Agrawal - San Jose CA
Roberto Javier Bayardo - San Jose CA
Dimitrios Gunopulos - Riverside CA
Ching-Tien Howard Ho - San Jose CA
Sunita Sarawagi - Maharashtra, IN
John Christopher Shafer - San Jose CA
Ramakrishnan Srikant - San Jose CA

International Business Machines Corporation - Armonk NY

G06F 1730

715513, 715 5, 715 10

A user can easily organize computerized document folders by associating a few sample documents in the document database with each folder. The present invention learns folder profiles based on the sample documents and moves the remaining documents into the folders accordingly. In this way, the user can construct new folders, or rearrange existing folders, or cause the computer to automatically rearrange and maintain the folders. This is particularly useful for managing a database of perhaps thousands of emails.

6633885, Oct 14, 2003

Jan 4, 2000

09/477257

Rakesh Agrawal - San Jose CA
John Christopher Shafer - San Jose CA

International Business Machines Corporation - Armonk NY

G06F 1730

707102, 707 3, 707 5, 705 1, 345764

A system and method for exporing a web-accessible database includes providing a GUI that a user can manipulate to quickly modify the results of a query to expand or contract the results set, without requiring additional querying. Attribute controls can be manipulated to impose restrictions on the results set, including by designating example records the attributes of which are used to restrict the records displayed to the user. Only records that can be displayed are instantiated, to further increase the speed of the system.

6651048, Nov 18, 2003

Oct 22, 1999

09/422657

Rakesh Agrawal - San Jose CA
Roberto J. Bayardo - San Jose CA

International Business Machines Corporation - Armonk NY

G06F 1730

707 2, 707 4, 707 6, 707102, 707104, 706 1

A method and structure for identifying database association rules includes mining first database association rules, the first database association rules having ratings with respect to a plurality of metrics and population constraints, selecting second database association rules from the first database association rules, each of the second database association rules having a highest rating with respect to a different metric of the metrics, and interactively changing the metrics and repeating the selecting to identify most important ones of the databases association rules for a given set of metrics.

6651049, Nov 18, 2003

Oct 22, 1999

09/422661

Rakesh Agrawal - San Jose CA
Roberto J. Bayardo - San Jose CA

International Business Machines Corporation - Armonk NY

G06F 1730

707 2, 707 4, 707 6, 707102, 707104, 706 1

A method and structure for identifying database association rules includes mining first database association rules, the first database association rules having ratings with respect to a plurality of metrics, selecting second database association rules from the first database association rules, each of the second database association rules having a highest rating with respect to a different metric of the metrics, and interactively changing the metrics and repeating the selecting to identify most important ones of the databases association rules for a given set of metrics.

6687691, Feb 3, 2004

Jan 19, 2000

09/487642

Rakesh Agrawal - San Jose CA
Ramakrishnan Srikant - San Jose CA

International Business Machines Corporation - Armonk NY

G06F 1730

707 6, 707 10, 707101

A system and method for mining data while preserving a users privacy includes perturbing user-related information at the users computer and sending the perturbed data to a Web site. At the Web site, perturbed data from many users is aggregated, and from the distribution of the perturbed data, the distribution of the original data is reconstructed, although individual records cannot be reconstructed. Based on the reconstructed distribution, a decision tree classification model or a Naive Bayes classification model is developed, with the model then being provided back to the users, who can use the model on their individual data to generate classifications that are then sent back to the Web site such that the Web site can display a page appropriately configured for the users classification. Or, the classification model need not be provided to users, but the Web site can use the model to, e. g. , send search results and a ranking model to a user, with the ranking model being used at the user computer to rank the search results based on the users individual classification data.

6687705, Feb 3, 2004

Jan 8, 2001

09/757047

Rakesh Agrawal - San Jose CA
Ramakrishnan Srikant - San Jose CA

International Business Machines Corporation - Armonk NY

G06F 1730

707101, 707 6, 707 7

A system and method for merging product information from a first hierarchy into a second hierarchy. A Naive Bayes classification model is generated using both text data and attribute (numerical) data pertaining to products in the second hierarchy. Then, products in the first hierarchy are placed into the second hierarchy in accordance with the model. Preferably, the placement of the products in the second hierarchy depends in part on their grouping in the first hierarchy, on the intuition that if two products were grouped together in the first hierarchy they have a higher likelihood of being grouped together in the second hierarchy as well.