Microsoft Technology Licensing, LLC

Brian Dean Haslam

Jan-Kristian Markiewicz

Arul A Menezes

Mitica Manu

Bradley A Music

Patrick Wayne John Evans

Patrick Santos

A computing system may predict a word based on received user input that selects a part of the word (e.g., the first characters, the first root, etc.). Specifically, a program, when run on the computing system, may perform a method including creating a candidate list of words based on received user input. These words may be then organized into a hierarchy, or tree structure, in which each word is associated with a parent and each parent is a partial match for its associated words. The top-tier partial matches may be presented, and user input corresponding to a selected partial match may be received. A set of candidates related to the selected partial match may then be presented for user selection.

Text prediction with partial selection in a variety of domains

The present invention may be used to analyze subject content, search and analyze reference content, compare the subject and reference content for similarity, and output comparison reports between the subject and reference content. The present invention may incorporate and utilize text from intrinsic and/or extrinsic subject documents. The analysis may employ a variety of metrics, including scores generated from a natural language processing system, scores based on classification similarity, scores based on proximity similarity, and in the case of analysis of patent documents, scores based on measurement of claims.

Parsing, analysis and scoring of document content

Growing community of inventors

North Bend, WA, United States of America

Brian Dean Haslam