Computer-readable recording medium storing information processing program, information processing apparatus, and information processing method

Patent metadata

Field	Value
Publication number	US-2024394733-A1
Application number	US-202418791619-A
Country	US
Kind code	A1
Filing date	Aug 1, 2024
Priority date	Feb 10, 2022
Publication date	Nov 28, 2024
Grant date	—

Abstract

A recording medium stores a program for causing a computer to execute a process including: referring to a memory storing data constituted by combinations of features to extract data groups of which the combinations satisfy each condition; identifying relationships between the features included in the data groups; classifying the relationships into a first clusters, based on first similarity; classifying the data groups into second clusters, based on second similarity; classifying the data groups into third clusters so as to classify, into a same cluster, data groups that are in a same one first cluster obtained by classifying the relationships corresponding correspond to each data group and are in a same one second cluster obtained by classifying each data group; identifying first conditions for classifying the data groups classified into each cluster and the data groups classified into other clusters; and outputting the identified first conditions with a classification result.

First claim

What is claimed is:

1 . A non-transitory computer-readable recording medium stores an information processing program for causing a computer to execute a process including:
referring to a memory that stores a plurality of pieces of data constituted by combinations of a plurality of features to extract, for each of a plurality of conditions, data groups of which the combinations satisfy each of the conditions; identifying, for each of the plurality of conditions, relationships between the plurality of features included in the data groups that correspond to each of the conditions; classifying the relationships for each of the plurality of conditions into a plurality of first clusters, based on first similarity between the relationships for each of the plurality of conditions; classifying the data groups for each of the plurality of conditions into a plurality of second clusters, based on second similarity between the data groups for each of the plurality of conditions; classifying the data groups for each of the plurality of conditions into a plurality of third clusters so as to classify, into a same cluster, a plurality of the data groups that are in a same one of the first clusters obtained by classifying the relationships that correspond to each of the data groups and are in a same one of the second clusters obtained by classifying each of the data groups; identifying, for each of the plurality of third clusters, first conditions capable of classifying the data groups classified into each cluster and the data groups classified into other clusters; and outputting the identified first conditions together with a classification result for the plurality of third clusters.

2 . The non-transitory computer-readable recording medium according to claim 1 , wherein
the identifying the first conditions includes identifying and outputting the first conditions for each cluster that has a predetermined number or more of the data groups classified into each cluster, among the plurality of third clusters.

3 . The non-transitory computer-readable recording medium according to claim 1 , wherein
the identifying the first conditions includes: identifying, for each of the plurality of third clusters, common data groups common to one or more of the data groups classified into each cluster; and identifying, for each of the plurality of third clusters, the conditions capable of classifying the common data groups classified into each cluster and the common data groups classified into other clusters, as the first conditions.

4 . The non-transitory computer-readable recording medium according to claim 3 , wherein
the identifying the first conditions includes: generating, for each of the plurality of third clusters, training models by performing machine learning with the common data groups classified into each cluster as positive examples and the common data groups classified into other clusters as negative examples; and identifying, for each of the plurality of third clusters, the conditions indicated by the training models that correspond to each cluster, as the first conditions.

5 . The non-transitory computer-readable recording medium according to claim 1 , wherein
the identifying the first conditions includes identifying, for each of the plurality of third clusters, common relationships common to one or more of the relationships that correspond to one or more of the data groups classified into each cluster, and the outputting includes outputting information that indicates the common relationships, as the classification result for the plurality of third clusters.

6 . The non-transitory computer-readable recording medium according to claim 5 , wherein
the outputting includes outputting information that indicates the relationships that are not included in the relationships between the plurality of features included in each piece of data stored in the memory, among the common relationships.

7 . An information processing apparatus comprising:
a memory; and a processor coupled to the memory and configured to: by referring to the memory that stores a plurality of pieces of data constituted by combinations of a plurality of features extract, for each of a plurality of conditions, data groups of which the combinations satisfy each of the conditions; identify, for each of the plurality of conditions, relationships between the plurality of features included in the data groups that correspond to each of the conditions; classify the relationships for each of the plurality of conditions into a plurality of first clusters, based on first similarity between the relationships for each of the plurality of conditions; classify the data groups for each of the plurality of conditions into a plurality of second clusters, based on second similarity between the data groups for each of the plurality of conditions; classify the data groups for each of the plurality of conditions into a plurality of third clusters so as to classify, into a same cluster, a plurality of the data groups that are in a same one of the first clusters obtained by classifying the relationships that correspond to each of the data groups and are in a same one of the second clusters obtained by classifying each of the data groups; identify, for each of the plurality of third clusters, first conditions capable of classifying the data groups classified into each cluster and the data groups classified into other clusters; and output the identified first conditions together with a classification result for the plurality of third clusters.

8 . The information processing apparatus according to claim 7 , wherein
the processor: identifies, for each of the plurality of third clusters, common data groups common to one or more of the data groups classified into each cluster; and identifies, for each of the plurality of third clusters, the conditions capable of classifying the common data groups classified into each cluster and the common data groups classified into other clusters, as the first conditions.

9 . The information processing apparatus according to claim 8 , wherein
the processor: generates, for each of the plurality of third clusters, training models by performing machine learning with the common data groups classified into each cluster as positive examples and the common data groups classified into other clusters as negative examples; and identifies, for each of the plurality of third clusters, the conditions indicated by the training models that correspond to each cluster, as the first conditions.

10 . An information processing method comprising:
referring to a memory that stores a plurality of pieces of data constituted by combinations of a plurality of features to extract, for each of a plurality of conditions, data groups of which the combinations satisfy each of the conditions; identifying, for each of the plurality of conditions, relationships between the plurality of features included in the data groups that correspond to each of the conditions; classifying the relationships for each of the plurality of conditions into a plurality of first clusters, based on first similarity between the relationships for each of the plurality of conditions; classifying the data groups for each of the plurality of conditions into a plurality of second clusters, based on second similarity between the data groups for each of the plurality of conditions; classifying the data groups for each of the plurality of conditions into a plurality of third clusters so as to classify, into a same cluster, a plurality of the data groups that are in a same one of the first clusters obtained by classifying the relationships that correspond to each of th

Assignees

• Fujitsu Ltd

Inventors

• Fujishige Yuta
• Asai Tatsuya
• UEMURA Kento
• Suzuki Hirofumi
• Koyanagi Yusuke
• Maruhashi Koji

Classifications

• G06N5/01

  Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound · CPC title

• G06N5/022

  Knowledge engineering; Knowledge acquisition · CPC title

• G06N20/00

  Machine learning · CPC title

• G06Q30/0201 Primary

  Market modelling; Market analysis; Collecting market data · CPC title