Dialog mechanism responsive to query context

• US 10,740,373 B2
• Filed: 02/08/2017
• Issued: 08/11/2020
• Est. Priority Date: 02/08/2017
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Altaf, Faheem, DeLuca, Lisa Seacat, Srinivas, Raghuram, Ugrani, Rajendra G.
• Primary Examiner(s): Le, Miranda
• Application Number:
  US15/427,109
• Publication Number:
  US 20180225365A1
• Time in Patent Office: 1,280 Days
• CPC Class Codes: G06F 16/334 Query execution G06F16/335 takes precedence
• CPC Class Codes: G06F 16/3344 using natural language analysis
• CPC Class Codes: G06F 16/358 Browsing; Visualisation therefor
• CPC Class Codes: G06F 16/90332 Natural language query formulation or dialogue systems
• CPC Class Codes: G06F 40/20 Natural language analysis semantic analysis of natural language G06F40/30

Abstract

Aspects automatically invoke automated dialog chat bots in response to determining that query text content meets a threshold relevance. A computer processor analyzes different words within text content of a user query input to identify via natural language processing query topics defined by combinations of the words. The processor drives a display device to present an automated dialog chat bot to the user that presents a chat bot series of dialog questions in response to determining that a query topic identified from analyzing the text content meets a threshold relevancy; or searches a database for results that satisfy keywords or query topics of the query content in response to failing to identify a query topic for the query text content, or to determining that an identified query topic does not meet the threshold relevancy.

Claims

1. A computer-implemented method for automatically invoking an automated dialog chat bot, comprising executing on a computer processor the steps of:

in response to an input of a user query text content comprising a plurality of different words, analyzing the text content to identify via natural language processing query topics defined by combinations of words and a sentence structure of the text content, comprising making query topic identifications from a limited set of chat bot topics that are each associated with products of a service provider;

in response to determining that a query topic is identified from the analyzing the text content, determining whether the identified query topic likely matches each of a plurality of different chat bot topics that are associated one each with each of different series of dialog questions;

ranking each of a plurality of ones of the different chat bot topics that are determined to likely match the identified query topic as a function of different relative values of strength of match of the identified query topic to the ones of the different chat bot topics and to service provider product values that are associated to the ones of the different chat bot topics, wherein the different relative values of strength of match are weighted to more highly rank the ones of the different chat bot topics that are associated to higher service provider product values;

determining that a highest ranked one of the ranked chat bot topics meets a threshold likelihood of matching chat bot topics defined for invoking an automated dialog chat bot;

searching a database for results that satisfy keywords in the query input text content or the query topics identified for the query input text content, in response to failing to identify a query topic from analyzing the text content via natural language processing, or in response to determining that the identified query topic does not likely match any of the plurality of different chat bot topics associated with the different series of dialog questions;

in response to determining that the highest ranked one of the ranked chat bot topics meets the threshold likelihood of matching chat bot topics, presenting an automated dialog chat bot on a display device within a graphical user interface structure; and

presenting in series from the automated dialog chat bot a plurality of dialog questions comprising ones of the series of dialog questions associated to the highest ranked one of the ranked chat bot topics and that are selected in response to answer inputs via the graphical user interface to a previous question in the presented series of dialog questions.

2. The method of claim 1, wherein the determining that the highest ranked one of the ranked chat bot topics meets the threshold likelihood comprises a step that is selected from the group consisting of:

determining whether time data of the input of the input query text content is within a promotional sales event period for a service provider product, wherein the time data of the input is selected from the group consisting of;

a time of day, and a calendar date;

determining whether a geographic location of the user is located within a geographical region that is associated with a geographic region-dependent value for a service provider product, wherein the geographic location of the user is selected from the group consisting of;

a home address of the user, and a business address of the user; and

determining whether the geographic location of the user meets a threshold qualification for matching the chat bot topic as a function of comparing the time data of the input of the user query text content to a time period of a disqualifying event, wherein the geographic location of the user is selected from the group consisting of;

a home address of the user, and a business address of the user.

3. The method of claim 1, wherein the series of dialog questions target aspects of a product of a service provider as a solution to the identified query topic.

4. The method of claim 1, wherein the step of analyzing the text content to identify query topics defined by combinations of words and the sentence structure of the text content further comprises:

making the query topic identifications from a limited set of chat bot topics that are each associated with products of a service provider.

5. A system, comprising:

a processor;

a computer readable memory in circuit communication with the processor; and

a computer readable storage medium in circuit communication with the processor;

wherein the processor executes program instructions stored on the computer-readable storage medium via the computer readable memory and thereby;

in response to an input of a user query text content comprising a plurality of different words, analyzes the text content to identify query topics via natural language processing that are defined by combinations of words and a sentence structure of the text content, comprising making query topic identifications from a limited set of chat bot topics that are each associated with products of a service provider;

in response to determining that a query topic is identified from the analyzing the text content, determines whether the identified query topic likely matches each of a plurality of different chat bot topics that are associated one each with each of different series of dialog questions;

ranks each of a plurality of ones of the different chat bot topics that are determined to likely match the identified query topic as a function of different relative values of strength of match of the identified query topic to the ones of the different chat bot topics and to service provider product values that are associated to the ones of the different chat bot topics, wherein the different relative values of strength of match are weighted to more highly rank the ones of the different chat bot topics that are associated to higher service provider product values;

determines that a highest ranked one of the ranked chat bot topics meets a threshold likelihood of matching chat bot topics defined for invoking an automated dialog chat bot;

searches a database for results that satisfy keywords in the query input text content or the query topics identified for the query input text content, in response to failing to identify a query topic from analyzing the text content via natural language processing, or in response to determining that the identified query topic does not likely match any of the plurality of different chat bot topics associated with the different series of dialog questions;

in response to determining that the highest ranked one of the ranked chat bot topics meets the threshold likelihood of matching chat bot topics, presents an automated dialog chat bot on a display device within a graphical user interface structure; and

presents in series from the automated dialog chat bot a plurality of dialog questions comprising ones of the series of dialog questions associated to the highest ranked one of the ranked chat bot topics and that are selected in response to answer inputs via the graphical user interface to a previous question in the presented chat bot series of dialog questions.

6. The system of claim 5, wherein the processor executes the program instructions stored on the computer-readable storage medium via the computer readable memory and thereby determines that the highest ranked one of the ranked chat bot topics meets the threshold likelihood via a process selected from the group consisting of:

determining whether time data of the input of the input query text content falls within a promotional sales event period for a time-dependent value of a service provider product, wherein the time data of the input is selected from the group consisting of;

a time of day, and a calendar date;

determining whether a geographic location of the user is located within a geographical region that is associated with a geographic region-dependent value for a service provider product, wherein the geographic location of the user is selected from the group consisting of;

a home address of the user, and a business address of the user; and

determining whether the geographic location of the user meets a threshold qualification for matching the chat bot topic as a function of comparing the time data of the input of the user query text content to a time period of a disqualifying event, wherein the geographic location of the user is selected from the group consisting of;

a home address of the user, and a business address of the user.

7. A computer program product for automatically invoking an automated dialog chat bot in response to query text content meeting a threshold relevance, the computer program product comprising:

a computer readable storage medium having computer readable program code embodied therewith, wherein the computer readable storage medium is not a transitory signal per se, the computer readable program code comprising instructions for execution by a processor that cause the processor to;

in response to an input of a user query text content comprising a plurality of different words, analyze the text content to identify via natural language processing query topics defined by combinations of words and a sentence structure of the text content, comprising making query topic identifications from a limited set of chat bot topics that are each associated with products of a service provider;

in response to determining that a query topic is identified from the analyzing the text content, determine whether the identified query topic likely matches each of a plurality of different chat bot topics that are associated one each with each of different series of dialog questions;

rank each of a plurality of ones of the different chat bot topics that are determined to likely match the identified query topic as a function of different relative values of strength of match of the identified query topic to the ones of the different chat bot topics and to service provider product values that are associated to the ones of the different chat bot topics, wherein the different relative values of strength of match are weighted to more highly rank the ones of the different chat bot topics that are associated to higher service provider product values;

determine that a highest ranked one of the ranked chat bot topics meets a threshold likelihood of matching chat bot topics defined for invoking an automated dialog chat bot;

search a database for results that satisfy keywords in the query input text content or the query topics identified for the query input text content, in response to failing to identify a query topic from analyzing the text content via natural language processing, or in response to determining that the identified query topic does not likely match any of the plurality of different chat bot topics associated with the different series of dialog questions

in response to determining that the highest ranked one of the ranked chat bot topics meets the threshold likelihood of matching chat bot topics, present an automated dialog chat bot on a display device within a graphical user interface structure; and

present in series a plurality of dialog questions comprising ones of the series of dialog questions associated to the highest ranked one of the ranked chat bot topics and that are selected in response to answer inputs from the graphical user interface to a previous question in the presented chat bot series of dialog questions.

8. The computer program product of claim 7, wherein the computer readable program code instructions for execution by the processor further cause the processor to determine that the highest ranked one of the ranked chat bot topics meets the threshold likelihood via a process selected from the group consisting of:

determining whether time data of the input of the input user query text content is within a promotional sales event period for a time-dependent value of a service provider product, wherein the time data of the input is selected from the group consisting of;

a time of day, and a calendar date;

determining whether a geographic location of the user is located within a geographical region that is associated with a geographic region-dependent value for a service provider product, wherein the geographic location of the user is selected from the group consisting of;

a home address of the user, and a business address of the user; and

determining whether the geographic location of the user meets a threshold qualification for matching the chat bot topic as a function of comparing the time data of the input of the user query text content to a time period of a disqualifying event, wherein the geographic location of the user is selected from the group consisting of;

a home address of the user, and a business address of the user.