Microsoft Technology Licensing, LLC

The Boeing Company

Nuance Communications, Inc.

Nobal Bikram Niraula

David Maxwell Chickering

Geoffrey G Zweig

Jason D Williams

Carlos Garcia Jurado Suarez

Aparna Lakshmiratan

Anne Kao

Andrey Kolobov

Daniel I Whyatt

Pradeep Dasigi

Amanda Joy Stent

I Dan Melamed

Hyuckchul Jung

A method is provided for analyzing and interpreting a dataset composed of electronic documents including free-form text. The method includes unifying terms of interest in the collection of terms of interest to identify variants of the terms of interest. This includes identifying candidate variants of a term of interest based on semantic similarity between the term of interest and other terms in the database, determined using an unsupervised machine learning algorithm. Linguistic features and contextual features of the term of interest and its candidate variants are extracted, at least the contextual features being extracted using the unsupervised machine learning algorithm. And a supervised machine learning algorithm is used with the linguistic features and contextual features to identify variants of the term of interest from the candidate variants, such as for application to generate features of the documents for data analytics performed thereon.

Unifying terms of interest from a dataset of electronic documents

Domain-specific language understanding models that may be built, tested and improved quickly and efficiently are provided. Methods, systems and devices are provided that enable a developer to build user intent detection models, language entity extraction models, and language entity resolution models quickly and without specialized machine learning knowledge. These models may be built and implemented via single model systems that enable the models to be built in isolation or in an end-to-end pipeline system that enables the models to be built and improved in a simultaneous manner.

Generation of language understanding systems and methods

A method is provided for analyzing and interpreting a dataset composed of electronic documents including free-form text. The method includes text mining the documents for terms of interest, including receiving a set of seed nouns as input to an iterative process an iteration of which includes searching for multiword terms having seed nouns as their head words, at least some of which define a training set of a machine learning algorithm used to identify additional multiword terms at least some of which have nouns outside the set of seed nouns as their head words. The iteration also includes adding the nouns outside the set of seed nouns to the set and thereby identifying a new set of seed nouns for a next iteration. The method includes unifying terms of interest to produce normalized terms of interest for application to generate features of the documents for data analytics performed thereon.

Text mining a dataset of electronic documents to discover terms of interest

Voice enabled dialog with web pages is provided. An Internet address of a web page is received including an area with which a user of a client device can specify information. The web page is loaded using the received Internet address of the web page. A task structure of the web page is then extracted. An abstract representation of the web is then generated. A dialog script, based on the abstract representation of the web page is then provided. Spoken information received from the user is converted into text and the converted text is inserted into the area.

Voice-enabled dialog interaction with web pages

Growing community of inventors

Memphis, TN, United States of America

Nobal Bikram Niraula