Answer sequence evaluation

Patent metadata

Field	Value
Publication number	US-2016125750-A1
Application number	US-201514708510-A
Country	US
Kind code	A1
Filing date	May 11, 2015
Priority date	Nov 5, 2014
Publication date	May 5, 2016
Grant date	—

Abstract

Aspects of the present disclosure are directed toward evaluating an answer sequence. Aspects are directed toward receiving a set of answer sequences including a first answer sequence. The first answer sequence may have a first set of answers. Aspects are also directed toward identifying a set of scores coupled with the first set of answers. Aspects are also directed toward determining, based on a subject matter corresponding to the first answer sequence, a set of evaluation rules. Aspects are also directed toward generating, based on the set of scores and the set of evaluation rules, a sequence evaluation score for the first answer sequence.

First claim

What is claimed is:

1 . A computer implemented method for evaluating an answer sequence comprising:
receiving a set of answer sequences including a first answer sequence having a first set of answers; identifying a set of scores coupled with the first set of answers; determining, based on a subject matter corresponding to the first answer sequence, a set of evaluation rules; and generating, based on the set of scores and the set of evaluation rules, a sequence evaluation score for the first answer sequence.

2 . The method of claim 1 , wherein determining the set of evaluation rules includes:
computing, based on the subject matter, a caution value for the first answer sequence; determining, by comparing the caution value to a first caution threshold, that the caution value achieves the first caution threshold; and selecting, in response to determining that the caution value achieves the first caution threshold, a first evaluation rule.

3 . The method of claim 2 , wherein:
the first evaluation rule identifies, from the set of scores coupled with the first set of answers, a first score of the set of scores, wherein the first score does not achieve a first score threshold; and generating the sequence evaluation score for the first answer sequence includes assigning, to the first answer sequence, the first score as the sequence evaluation score.

4 . The method of claim 2 , further comprising:
determining, by comparing the caution value to a second caution threshold, that the caution value does not achieve the second caution threshold; selecting, in response to determining that the caution value does not achieve the second caution threshold, a second evaluation rule.

5 . The method of claim 4 , wherein:
the second evaluation rule calculates, by performing a statistical algorithm on the set of scores, an aggregate score; and generating the sequence evaluation score for the first answer sequence comprises assigning, to the first answer sequence, the aggregate score as the sequence evaluation score.

6 . The method of claim 1 , further comprising:
identifying a set of answer categories corresponding to the set of answers of the first answer sequence; and collecting, for the set of categories corresponding to the set of answers of the first answer sequence, context data indicating a relative importance of a first answer category of the set of answer categories to the first answer sequence.

7 . The method of claim 6 , wherein determining the set of evaluation rules includes:
evaluating, using a natural language processing technique configured to parse semantic and syntactic content, the context data; determining, in response to evaluating the context data, that the context data achieves a satisfaction criterion; and selecting, in response to determining that the context data achieves the satisfaction criterion, a third evaluation rule.

8 . The method of claim 6 , wherein:
a third evaluation rule further comprises:
assigning, based on the context data, a first weighting value to the first answer category of the set of answer categories,
assigning, based on the context data, a second weighting value to a second answer category of the set of answer categories, and
calculating, by a statistical algorithm using the first weighting value and the second weighting value, an aggregate score; and
generating the sequence evaluation score for the first answer sequence includes assigning, to the first answer sequence, the aggregate score as the sequence evaluation score.

9 . The method of claim 8 , further comprising:
receiving, from a user, a first set of answer preference data indicating an inclination for the first answer category of the set of categories; and increasing, based on the first set of answer preference data, the first weighting value assigned to the first answer category.

10 . The method of claim 8 , further comprising:
receiving, from a user, a second set of answer preference data indicating a disinclination for the second answer category of the set of categories; and decreasing, based on the second set of answer preference data, the second weighting value assigned to the second answer category.

11 . The method of claim 1 , further comprising:
comparing the first answer sequence with a second answer sequence; identifying, based on comparing the first answer sequence with the second answer sequence, a first answer category that belongs to the first answer sequence and is absent from the second answer sequence; determining, in response to identifying the first answer category, that a first score coupled with a first answer of the first answer category achieves a first influence threshold; and modifying, in response to determining that the first score achieves the first influence threshold, the sequence evaluation score of the second answer sequence.

12 . The method of claim 1 , further comprising filtering, from the first answer sequence, a second answer category including a second set of answers, wherein the second set of answers fail to achieve a second score threshold.

Assignees

• IBM

Inventors

• Barker Kevin S
• Eggebraaten Thomas J
• Megerian Mark G

Classifications

• G06N7/01

  Probabilistic graphical models, e.g. probabilistic networks · CPC title

• G06Q30/0203

  Market surveys; Market polls · CPC title

• G16H50/70

  for mining of medical data, e.g. analysing previous cases of other patients · CPC title

• G06F40/30

  Semantic analysis · CPC title

• G06F16/90324

  using system suggestions · CPC title