FREQUENT PATTERN EXTRACTION APPARATUS FREQUENT PATTERN EXTRACTION METHOD AND PROGRAM

• US 20130179381A1
• Filed: 01/03/2013
• Published: 07/11/2013
• Est. Priority Date: 01/11/2012
• Status: Active Grant

• Current Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Original Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Inventors: Kawabata, Takayuki, Itoh, Fumiaki, Yokota, Haruo, Watanabe, Yosuke, Song, Qiang
• Granted Patent: US 9,262,714 B2
• Time in Patent Office: Days
• US Class Current: 706/46
• CPC Class Codes: G06F 16/1734 Details of monitoring file system events, e.g. by the use of hooks, filter drivers, logs
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation

Abstract

There is provided a frequent pattern extraction apparatus. In the apparatus, time series data for an item operation is divided into a plurality of item operation sets. Using a set degree of similarity between the item operation sets, an abstract item operation set is generated. A pattern of sequences that frequently appear is extracted from the sequences in the abstract item operation set. Using the pattern of sequences that frequently appear, an item as well as an item operation are recommended to a user.

Claims

1. A frequent pattern extraction apparatus comprising:

an acquisition unit configured to acquire time series data for a data operation in user units from a data operation history database;

a dividing unit configured to divide the time series data for the data operation into a plurality of data operation sets;

a set degree of similarity acquisition unit configured to acquire a set degree of similarity among the plurality of data operation sets;

a generation unit configured to generate an abstract data operation set from the plurality of data operation sets using the set degree of similarity; and

an extraction unit configured to extract a pattern of sequences that frequently appear among sequences in the abstract data operation set.

2. The frequent pattern extraction apparatus according to claim 1, wherein the dividing unit is configured to divide the time series data for the data operation into the plurality of data operation sets at fixed time intervals.

3. The frequent pattern extraction apparatus according to claim 1, wherein the dividing unit is configured to divide the time series data for the data operation into the plurality of data operation sets at locations where a time between data operations is equal to or more than a fixed duration.

4. The frequent pattern extraction apparatus according to claim 1, further comprising:

a data degree of similarity acquisition unit configured to calculate a data degree of similarity among the data; and

a clustering unit configured to cluster data using the data degree of similarity,

wherein the plurality of data operation sets are clustered into cluster units by the clustering unit.

5. The frequent pattern extraction apparatus according to claim 4, wherein the data degree of similarity is acquired using a copy history among the data.

6. The frequent pattern extraction apparatus according to claim 4, wherein the data degree of similarity is acquired using a data name of the data.

7. The frequent pattern extraction apparatus according to claim 1, further comprising:

a determination unit configured to determine the abstract data operation set similar to a data operation set obtained by dividing time series data for a most recent data operation by a user by using the set degree of similarity; and

a recommendation unit configured to recommend data to the user by using a pattern of sequences to which the determined abstract data operation set belongs.

8. The frequent pattern extraction apparatus according to claim 7, wherein a data operation is recommended by the recommendation unit.

9. A method for extracting a frequent pattern comprising:

acquiring time series data for a data operation in user units from a data operation history database;

dividing the time series data for the data operation into a plurality of data operation sets;

acquiring a set degree of similarity among the data operation sets;

generating an abstract data operation set from the plurality of data operation sets using the set degree of similarity; and

extracting a pattern of sequences that frequently appear among sequences in the abstract data operation set.

10. A non-transitory storage medium in which a computer program is stored, the program, when executed by a computer, causes the computer to perform a method for extracting a frequent pattern, the method comprising:

acquiring time series data for a data operation in user units from a data operation history database;

dividing the time series data for the data operation into a plurality of data operation sets;

acquire a set degree of similarity among the plurality of data operation sets;

generating an abstract data operation set from the data operation sets using the set degree of similarity; and

extracting a pattern of sequences that frequently appear among sequences in the abstract data operation set.