OUTBOUND CONTENT FILTERING VIA AUTOMATED INFERENCE DETECTION

Patent №

US 8,990,225

Granted

2015-03-24

Filed 2007

Owner

PALO ALTO RESEARCH CENTER, INC.

Lab

AI components

4

ml · nlp · kr · planning

Assignment

Recorded

Dataset

AIPD

2023_r1 edition

Application

11957833

One embodiment of the present invention provides a system that facilitates filtering outbound content via inference detection. During operation, the system identifies content sent to a first address and extracts keywords from the identified content. The system then issues queries based on these keywords and extracts expected-content keywords from the hits returned in response to the queries. The system then searches the outbound content for occurrences of the expected-content keywords and produces a result which allows a user to determine whether the outbound content is proper. In a further embodiment, the system extracts keywords from a piece of outbound content, and issues queries based on these keywords. The system then extracts keywords from the hits, and present at least one keyword to a user, thereby allowing the user to determine whether the outbound content is proper.

AI classification

Natural language1.00
Planning1.00
Knowledge representation0.96
Machine learning0.94
AI hardware0.48
Speech0.13
Vision0.00
Evolutionary computation0.00

Ownership

PALO ALTO RESEARCH CENTER, INC.

assignment · 202570583

Assignors

CHOW, RICHARD, GOLLE, PHILIPPE J.P., STADDON, JESSICA N.

On an employer assignment, the assignors are typically the inventors.

From the same owner

© 2026 NYSGPT2525 LLC