Visual representation of question quality

• US 9,910,886 B2
• Filed: 04/17/2015
• Issued: 03/06/2018
• Est. Priority Date: 04/17/2015
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Adams, Jr., Richard H., Chandrasekaran, Swaminathan, Sudarsan, Sridhar, Taieb, David D.
• Primary Examiner(s): Pham, Tuan A
• Application Number:
  US14/689,106
• Publication Number:
  US 20160306846A1
• Time in Patent Office: 1,054 Days
• CPC Class Codes: G06F 16/2423 Interactive query statement specification based on a database schema
• CPC Class Codes: G06F 16/243 Natural language query formulation
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation

Abstract

Aspects of the present disclosure, in certain embodiments, are directed toward providing a visual representation of question quality in a question answering system. More particularly, aspects are directed toward receiving, by a question answering system, an input question having a set of query attributes. The set of query attributes may be user-influenced characteristics of the input question. Aspects of the present disclosure are also directed toward evaluating, by comparing the set of query attributes to a set of assessment criteria, the quality of the input question. Aspects of the present disclosure are also directed toward assigning, in response to evaluating the quality of the input question, a set of quality values to the set of query attributes. Using the assigned set of quality values and the set of query attributes, an icon that indicates a visual representation of the quality of the input question may be generated.

Claims

1. A method in a deep question answering system for providing a visual representation of question quality, the method comprising:

receiving, by a question answering system, an input question having a set of query attributes, wherein the set of query attributes are user-influenced characteristics of the input question;

evaluating, by comparing the set of query attributes to a set of assessment criteria, the quality of the input question, wherein evaluating includes;

parsing, using a natural language processing technique configured to analyze semantic and syntactic content, the input question,

identifying, based on the semantic and syntactic content, the set of query attributes of the input question, and

determining whether each query attribute of the set of query attributes achieves a corresponding satisfaction threshold of a set of satisfaction thresholds with respect to the set of assessment criterion;

assigning, in response to evaluating the quality of the input question, a respective quality value of a set of quality values to each query attribute of the set of query attributes, wherein each respective quality value is a value that indicates whether the respective query attribute satisfies the corresponding satisfaction threshold; and

generating, using the assigned set of quality values and the set of query attributes, an icon that indicates a visual representation of the quality of the input question;

wherein generating the icon comprises;

mapping each query attribute of the set of query attributes to a respective element of the icon; and

selecting a corresponding set of visual characteristics for each respective element of the icon based on the respective quality value assigned to the query attributed mapped to the respective element of the icon to visually indicate whether the query attribute mapped to the respective element of the icon satisfies the corresponding satisfaction threshold;

the method further comprising;

assigning, to a subset of the set of query attributes, a set of weighting values, wherein the set of weighting values indicates a quantitative representation of the relative importance of a particular query attribute of the set of query attributes with respect to the input question;

collecting, by analyzing the input question using a linguistic profiling technique, multi-factorial user context data including age data, level of education data, native language data, sentiment data, and input device data;

modifying, based on the multi-factorial user context data, the set of weighting values assigned to the set of query attributes; and

calculating, based on the set of weighting values assigned to the set of query attributes, the set of quality values.

2. The method of claim 1, wherein the set of query attributes include query syntax, part-of-speech usage, query length, and query complexity.

3. The method of claim 1, wherein the set of query attributes further include abbreviation usage, ambiguous terminology, missing elements, and question focus.

4. The method of claim 1, wherein generating the icon includes:

mapping, to a first facial element of a digital face icon, a first query attribute of the set of query attributes;

selecting, based on a first quality value assigned to the first query attribute, a set of visual characteristics for the first facial element.

5. The method of claim 4, wherein the set of visual characteristics for the first facial element include color, shape, size, location, and animation style.

6. The method of claim 4, further comprising:

assembling, based on the set of quality values, the set of weighting values, and the set of query attributes, a digital face icon that indicates a visual representation of the quality of the input question.

7. The method of claim 6, further comprising:

collecting, using a graphical user interface, user feedback data regarding the digital face icon;

defining, using a machine learning technique configured to process the user feedback data, a decision parameter for generating the digital face icon; and

adjusting, based on the feedback data and the decision parameter, the set of visual characteristics of the digital face icon.

8. A computer system comprising:

at least one processor;

a memory coupled to the at least one processor;

a visual representation generation mechanism executed by one or more of the at least one processor, the visual representation generation mechanism including;

a receiving module to receive, using a question answering system, an input question having a set of query attributes, wherein the set of query attributes are user-influenced characteristics of the input question;

an evaluating module to evaluate, by comparing the set of query attributes to a set of assessment criteria, the quality of the input question, wherein evaluating includes;

a parsing module to parse, using a natural language processing technique configured to analyze semantic and syntactic content, the input question,

an identifying module to identify, based on the semantic and syntactic content, the set of query attributes of the input question, and

a determining module to determine whether each query attribute of the set of query attributes achieves a corresponding satisfaction threshold of a set of satisfaction thresholds with respect to the set of assessment criterion;

an assigning module to assign, in response to evaluating the quality of the input question, a respective quality value of a set of quality values to each query attribute of the set of query attributes, wherein each respective quality value is a value that indicates whether the respective query attribute satisfies the corresponding satisfaction threshold; and

a generating module to generate, using the assigned set of quality values and the set of query attributes, an icon that indicates a visual representation of the quality of the input question, wherein the generating module is configured to generate the icon by;

mapping each query attribute of the set of query attributes to a respective element of the icon; and

selecting a corresponding set of visual characteristics for each respective element of the icon based on the respective quality value assigned to the query attributed mapped to the respective element of the icon to visually indicate whether the query attribute mapped to the respective element of the icon satisfies the corresponding satisfaction threshold;

wherein the processor is configured to;

assign, to a subset of the set of query attributes, a set of weighting values, wherein the set of weighting values indicates a quantitative representation of the relative importance of a particular query attribute of the set of query attributes with respect to the input question;

collect, by analyzing the input question using a linguistic profiling technique, multi-factorial user context data including age data, level of education data, native language data, sentiment data, and input device data;

modify, based on the multi-factorial user context data, the set of weighting values assigned to the set of query attributes; and

calculate, based on the set of weighting values assigned to the set of query attributes, the set of quality values.

9. The system of claim 8, wherein the set of query attributes include query syntax, part-of-speech usage, query length, and query complexity.

10. The system of claim 8, wherein the set of query attributes further include abbreviation usage, ambiguous terminology, missing elements, and question focus.

11. A computer program product comprising a non-transitory computer readable storage medium having a computer readable program stored therein, wherein the computer readable program, when executed on a first computing device, causes the computing device to:

receive, by a question answering system, an input question having a set of query attributes, wherein the set of query attributes are user-influenced characteristics of the input question;

evaluate, by comparing the set of query attributes to a set of assessment criteria, the quality of the input question, wherein being caused to evaluate includes being caused to;

parse, using a natural language processing technique configured to analyze semantic and syntactic content, the input question,

identify, based on the semantic and syntactic content, the set of query attributes of the input question, and

determine whether each query attribute of the set of query attributes achieves a corresponding satisfaction threshold of a set of satisfaction thresholds with respect to the set of assessment criterion;

assign, in response to evaluating the quality of the input question, a respective quality value of a set of quality values to each query attribute of the set of query attributes, wherein each respective quality value is a value that indicates whether the respective query attribute satisfies the corresponding satisfaction threshold; and

generate, using the assigned set of quality values and the set of query attributes, an icon that indicates a visual representation of the quality of the input question, wherein, to generate the icon, the computer readable program causes the computing device to;

map each query attribute of the set of query attributes to a respective element of the icon; and

select a corresponding set of visual characteristics for each respective element of the icon based on the respective quality value assigned to the query attributed mapped to the respective element of the icon to visually indicate whether the query attribute mapped to the respective element of the icon satisfies the corresponding satisfaction threshold;

wherein the computer readable program further causes the computing device to;

assign, to a subset of the set of query attributes, a set of weighting values, wherein the set of weighting values indicates a quantitative representation of the relative importance of a particular query attribute of the set of query attributes with respect to the input question;

collect, by analyzing the input question using a linguistic profiling technique, multi-factorial user context data including age data, level of education data, native language data, sentiment data, and input device data;

modify, based on the multi-factorial user context data, the set of weighting values assigned to the set of query attributes; and

calculate, based on the set of weighting values assigned to the set of query attributes, the set of quality values.

12. The computer program product of claim 11, wherein the set of query attributes include query syntax, part-of-speech usage, query length, query complexity, abbreviation usage, ambiguous terminology, missing elements, and question focus.

13. The computer program product of claim 11, wherein the computer readable program further causes the computing device to:

map, to a first facial element of a digital face icon, a first query attribute of the set of query attributes;

select, based on a first quality value assigned to the first query attribute, a set of visual characteristics for the first facial element.

14. The computer program product of claim 13, wherein the set of visual characteristics for the first facial element include color, shape, size, location, and animation style.

15. The computer program product of claim 13, wherein the computer readable program further causes the computing device to:

assemble, based on the set of quality values, the set of weighting values, and the set of query attributes, a digital face icon that indicates a visual representation of the quality of the input question.

16. The computer program product of claim 15, wherein the computer readable program further causes the computing device to:

collect, using a graphical user interface, user feedback data regarding the digital face icon;

define, using a machine learning technique configured to process the user feedback data, a decision parameter for generating the digital face icon; and

adjust, based on the feedback data and the decision parameter, the set of visual characteristics of the digital face icon.