Browse Prior Art Database

System and Method to Generate User Profiles Based On Question and Answer Interchange

IP.com Disclosure Number: IPCOM000238977D
Publication Date: 2014-Sep-29
Document File: 4 page(s) / 46K

Publishing Venue

The IP.com Prior Art Database

Abstract

A system and method to generate user profiles based on question and answer interchange is disclosed.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 28% of the total text.

Page 01 of 4

System and Method to Generate User Profiles Based On Question and Answer Interchange

Disclosed is a system and method to generate user profiles based on question and answer interchange.

The disclosed system and method classifies users under a set of predetermined profiles based on natural conversation interaction. This interchange can occur in many

ways, one of which is through use of a Question & Answer system. By better understanding what type of consumer is asking the questions, insights about the user and their motivation are derived. These insights are used to achieve higher user satisfaction and promote and more meaningful exchange of information. Today, there is no such way to glean this information other than through human intuition or reactively by analyzing past transactions or searches the consumer has made. The disclosed system is a real-time system that adapts with each conversational interchange.

People interact with knowledge sources to exchange information through different mediums everyday. The form of these interactions varies as well, and often times takes place as question and answer interaction. This form occurs through natural human conversation, but also through use of a Question & Answer system. When such dialogue exchanges occur between a firm and customer or potential customer, it is important to understand what type of person is asking the questions in order to better service their needs. The type of person asking the question can be a profile abstraction based on characteristics glean through the course of the conversation. Today, there is no such way to glean this information other than through human intuition and industry experience. This human element method works for person-to-person interaction, but cannot be recreated when the customer is interacting with a system, such as a question and answer system.

Converting user interaction with knowledge sources into characteristics about the user is crucial for gleaning key insights about user and their motivation in order to achieve higher user satisfaction and promote and more meaningful exchange of information. The disclosed system achieves the goal of understanding the end-user through a method of natural language generalization and characteristic assignment to assign the user to a predetermined profile based on an ongoing flow of conversation. The system utilizes lexical analysis type (LAT) detection, entity resolution, syntactic and semantic frames to derive domain defined generic concepts.

Two primary advantages are highlighted:
1) Dynamic Profiling - Today, a primary method of ascertaining information about a customer is through questionnaires. For example, firms within the financial industry may use an Investor Suitability Questionnaire. Through this method, a static set of questions is sent to the customer to answer. From these responses, the sending firm understands the customer's motivation and better serves them. Limitations exist which includ...