Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Unstructured APIs: User-Customized Unstructured API Broker

IP.com Disclosure Number: IPCOM000237212D
Publication Date: 2014-Jun-09
Document File: 2 page(s) / 25K

Publishing Venue

The IP.com Prior Art Database

Abstract

A method for providing unstructured Application Programming Interfaces (APIs) for: user customized unstructured API broker is disclosed.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 53% of the total text.

Page 01 of 2

Unstructured APIs: User-Customized Unstructured API Broker

Disclosed is a method for providing unstructured Application Programming Interfaces (APIs) for: user customized unstructured API broker.

It is desirable to use natural language as much as possible to interact with any service or features. Voice based processing systems provide services and answers by analyzing natural language. Providing an interface, a broker to the service, may be improved by using user context and user profile information to be more intelligent and intuitive when brokering a request. The user's profile, context and historical actions are utilized to intuitively choose the right broker for an ambiguous natural language request to an underlying API service providing the right service calls based on the users data.

Preferred Embodiment

The Unstructured API broker is configured to receive a users profile information, context and historical actions when being prompted to broker a request. The broker analyzes the question and finds the goal and key terms in natural language text, and finds similar goals and terms within the users profile and historical actions. The broker then queries to find services that match that goal and terms, then it finds the appropriate service that matches that specific users interests from profile or historical data. The service is then used to service the request.

Implementation Steps

A user asks for a service in natural language


1.

The question for the service is sent to the broker, including the context, user profile


2.

information and historical data


The broker analyzes the question for the goal and answer typ...