Browse Prior Art Database

System for generating a set of questions from a natural language document Disclosure Number: IPCOM000236082D
Publication Date: 2014-Apr-04
Document File: 2 page(s) / 76K

Publishing Venue

The Prior Art Database


Resolving issues associated with communication between two or more parties, where questions posed in the communication remain unanswered. The solution presented uses natural language processing to identify the questions within the communication and present them to the recipient(s) in a 'marked-up' format.

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

Page 01 of 2

System for generating a set of questions from a natural language document

In our busy lives we receive many different written requests for assistance, for example emails and instant messages. Due to the free-form nature of these communication methods, it can be easy to overlook some of the questions that are being asked and respond with an incomplete answer. Typically when we have asked a question to which an incomplete answer has been provided we then have to restate the question, or at least part of it, to receive a response.

    There are a variety of reasons why this can occur, including, but not limited to:

Lack of appropriate punctuation

Language barriers
Poor formatting of the questions

As a result we enter into a to and fro of requests and responses, wasting time

on both sides of the conversation, and in the case of time critical requests, money.

    This invention addresses this problem by processing the message content between the message being sent and subsequently received in order to identify all the questions within the request, presenting them to the recipient in such a way that it is obvious how many questions require answers, thereby minimising the time

wasted as a result of an incomplete response.

    For example, it would be feasible to create an email template that allows a sender to clearly enter a series of questions within a designated "question field", however this breaks the underlying nature of the free-form written communication, imposing restrictions on the sender.

    This idea introduces an intermediate step between the message sender and the receiver that processes the natural language in the message to identify each question contained within it. The processing step adds meta-data to the message to help the receiving application locate and present the questions in a suitable form to the receiver. For example, simply highlighting the sentences that represent questions.

    Natural language processing algorithms do not fall foul of the typical problems a human reader has, such as lack of time or lack of sufficient language skills, and are able to tolerate errors in punctuation, yet still determine if a sentence represents a question or a statement.

    Having processed the message, adding meta-data that identifies the location of the questions within it, the receiving application is then able to use this data to enhance the representation of the message, or en...