Analysis method, analysis apparatus, and non-transitory computer-readable storage medium for storing program

Patent metadata

Field	Value
Publication number	US-11367025-B2
Application number	US-201916375882-A
Country	US
Kind code	B2
Filing date	Apr 5, 2019
Priority date	Apr 26, 2018
Publication date	Jun 21, 2022
Grant date	Jun 21, 2022

Abstract

An analysis method includes: executing first accumulation processing for accumulating analysis reports including an analysis item regarding an analysis target and analysis results with respect to the analysis item; executing first extraction processing for extracting the analysis item and texts representing the analysis results from each of the analysis reports accumulated; executing first identification processing for identifying analysis techniques corresponding to the texts extracted; executing generation processing for generating analysis patterns; executing second identification processing for identifying first analysis patterns; executing third identification processing for identifying other analysis patterns; executing second accumulation processing for accumulating pattern information; executing second extraction processing for extracting the analysis item and the texts from a new analysis report; and executing output processing for identifying an analysis technique.

First claim

What is claimed is:

1. A computer implemented method of analysis, the method comprising:
executing, by one or more computer processors, first accumulation processing that includes accumulating analysis reports in a storage device, each of the analysis reports being a report generated by the one or more computer processors using any one of a plurality of predefined templates, each of the analysis reports including an analysis item regarding an analysis target and analysis results with respect to the analysis item;
executing, by the one or more computer processors, first extraction processing that includes extracting, from each of the analysis reports accumulated, the analysis item and texts, each of the extracted texts including a character sequence representing an individual result of the analysis results;
executing, by the one or more computer processors, first identification processing that includes
converting the texts extracted from each of the analysis reports into converted texts by replacing a part of each of the extracted texts with any one of a plurality of predetermined text labels each of which includes a predetermined wording, and
obtaining, for a respective text of the converted texts from each analysis report, an analysis technique corresponding to the respective text by using correspondence information indicating a relationship between each text of the converted texts and a corresponding analysis technique;
executing, by the one or more computer processors, generation processing that includes generating, for the respective text extracted from each analysis report, an analysis pattern including: the analysis item extracted, the respective text extracted, and the analysis technique identified, the generated analysis pattern being associated with the each analysis report;
executing, by the one or more computer processors, second identification processing that includes
calculating appearance frequency of each analysis technique in a plurality of analysis patterns obtained by the generation processing performed for the respective text of each analysis report,
obtaining, for each analysis technique in the plurality of analysis patterns, a determination result by determining whether the calculated appearance frequency of that analysis technique is equal to or lower than a threshold, and
in response to the determination result indicating that the calculated appearance frequency of that analysis technique is equal to or lower than the threshold, obtaining, as a first analysis pattern from among the plurality of analysis patterns an analysis pattern including that analysis technique;
executing, by the one or more computer processors, third identification processing that includes
obtaining other analysis patterns from each analysis report associated with the first analysis pattern, the other analysis patterns being analysis patterns associated with the each analysis report same as the first analysis pattern, the other analysis patterns being analysis patterns other than the first analysis pattern among analysis patterns associated with the each analysis report same as the first analysis pattern,
obtaining second analysis patterns from among the other analysis patterns, each of the second analysis patterns being an analysis pattern having a same pair of the analysis item and the text among analysis reports associated with the first analysis pattern, and
obtaining other analysis reports from among the accumulated analysis reports in the storage device, each of the other analysis reports being an analysis report that does not includes an analysis pattern same as the first analysis pattern;

in response that at least any one of the second analysis patterns is not included in each of the other analysis reports, executing, by the one or more computer processors, second accumulation processing that includes
generating pattern information indicating a relationship between the analysis technique corresponding to the first analysis patterns and the at least any one of the second analysis patters, and
accumulating the generated pattern information in the storage device;

executing, by the one or more computer processors, second extraction processing that includes extracting, in response to acceptance of a new analysis report, the analysis item and the texts from the new analysis report; and
executing, by the one or more computer processors, output processing that includes
identifying a new analysis technique corresponding to the second analysis patterns by using the pattern information accumulated in the storage device in case that the analysis item and the texts extracted from the new analysis report are respectively equal to the analysis item and the text included in each of the second analysis patterns, and
outputting an analysis result of the analysis target by performing the identified new analysis technique on that analysis target, to accumulate the new analysis report in the storage device in conjunction with the analysis result obtained by the performing of the identified new analysis technique.

2. The analysis method according to claim 1 , further comprising:
executing fourth identification processing that includes identifying third analysis reports from among the accumulated analysis reports, each of the third analysis reports being an analysis report that is not associated with the first analysis patterns, and
wherein the identifying of the second analysis patterns is configured to identify the second analysis patterns from among the other analysis patterns, each of the second analysis patterns being an analysis pattern having the same pair of the analysis item and the text among analysis reports associated with the first analysis pattern, the text included in each of the second analysis patterns being different from the text included in each analysis report associated with the identified third analysis report.

3. The analysis method according to claim 1 , wherein the second accumulation processing is configured to avoid accumulation of the pattern information including the first analysis pattern identified and the second analysis pattern identified when the pattern information including the first analysis pattern identified and the second analysis pattern identified has been already accumulated.

4. The analysis method according to claim 1 , wherein the second accumulation processing is configured to delete the pattern information that includes the first analysis pattern identified and does not include the second analysis pattern identified from the storage device when first pattern information and second pattern information have been already accumulated, the first pattern information being the pattern information that includes the first analysis pattern identified and the second analysis pattern identified, the second pattern information being the pattern information that includes the first analysis pattern identified and does not include the second analysis pattern identified.

5. The analysis method according to claim 1 , wherein the output processing is configured to
increase the number of times of generation of the pattern information corresponding to the second analysis pattern including the analysis item and the text extracted by a given number, and
delete, from the storage device, the pattern information other than the pattern information including the second analysis pattern regarding which the number of times of generation is largest in a plurality of pieces of the pattern information including the first analysis pattern identical when the plurality of pieces of the pattern information have been accumulated in the storage device.

6. The analysis method according to claim 1 , wherein the output processing is configured to
increase the number of times of generation o

Assignees

• Fujitsu Ltd

Inventors

• Uchiumi Tetsuya
• Saito Yuji
• Asaoka Masahiro
• Kondo Reiko
• Watanabe Yukihiro

Classifications

• G06V30/127

  with the intervention of an operator · CPC title

• G06F18/285

  Selection of pattern recognition techniques, e.g. of classifiers in a multi-classifier system · CPC title

• G06N5/027

  Frames · CPC title

• G06Q10/063 Primary

  Operations research, analysis or management · CPC title

• G06F40/279

  Recognition of textual entities · CPC title