Automatic feature extraction from a relational database

Patent metadata

Field	Value
Publication number	US-11200263-B2
Application number	US-201916600837-A
Country	US
Kind code	B2
Filing date	Oct 14, 2019
Priority date	Jan 17, 2017
Publication date	Dec 14, 2021
Grant date	Dec 14, 2021

Abstract

Techniques facilitating automatic feature extraction from a relational database are provided. In an embodiment, a method can include generating an entity graph based on a relational database, wherein the entity graph comprises a first node associated with a first table in the relational database and a second node associated with a second table in the relational database. In another embodiment, the method can include joining the first table and the second table based on an edge between the first table and the second table defined by the entity graph, wherein a resulting joined table is connected by a column of data. In another embodiment, the method can include extracting a feature from the column of data using a data mining algorithm selected from a set of data mining algorithms based on a type of data in the column of data.

First claim

What is claimed is:

1. A computer-implemented method, comprising:
generating, by a device operatively coupled to a processing unit, an entity graph with a first node representing a first table of a relational database and a second node representing a second table of the relational database;
forming, by the device, a joined table based on an edge between the first node and the second node, wherein the joined table is connected by a column of data; and
extracting, by the device using a data mining algorithm, one or more features from tables of the relational database based on the column of data of the joined table.

2. The computer-implemented method of claim 1 , further comprising:
selecting, by the device, the data mining algorithm from a set of data mining algorithms for the feature extraction based on determining a type of data in the column of data from a group consisting of spatial-temporal data, time-series data, sequence data, item set data, number set data, singleton data, text data, and image data.

3. The computer-implemented method of claim 1 , further comprising:
selecting, by the device, a feature from the one or more features based on a relevance to a target variable that is defined by an entity in a main table associated with a root node of the entity graph.

4. The computer-implemented method of claim 1 , further comprising:
collecting, by the device, the one or more features extracted from tables of the relational database by traversing the entity graph.

5. The computer-implemented method of claim 4 , wherein the entity graph is traversed to a depth based on a defined criterion related to processing efficiency.

6. The computer-implemented method of claim 4 , wherein the entity graph is traversed to a depth based on a defined criterion related to a user input.

7. The computer-implemented method of claim 4 , wherein the collecting further comprises:
transforming, by the device, a collection path into a canonical form; and
checking, by the device, an equivalent path to the canonical form to avoid redundant path traversal.

8. A system, comprising:
a memory that stores computer executable components; and
a processor that executes the computer executable components stored in the memory, wherein the computer executable components comprise:
a graphing component configured to generate an entity graph with a first node representing a first table of a relational database and a second node representing a second table of the relational database;
a joining component configured to form a joined table based on an edge between the first node and the second node, wherein the joined table is connected by a column of data; and
a feature extraction component configured to employ a data mining algorithm to extract one or more features from tables of the relational database based on the column of data of the joined table.

9. The system of claim 8 , further comprising:
a selection component configured to select the data mining algorithm from a set of data mining algorithms for the feature extraction based on determining a type of data in the column of data from a group consisting of spatial-temporal data, time-series data, sequence data, item set data, number set data, singleton data, text data, and image data.

10. The system of claim 8 , further comprising:
an identification component configured to select a feature from the one or more features based on a relevance to a target variable that is defined by an entity in a main table associated with a root node of the entity graph.

11. The system of claim 8 , further comprising:
a collection component configured to collect the one or more features extracted from tables of the relational database by traversing the entity graph.

12. The system of claim 11 , wherein the entity graph is traversed to a depth based on a defined criterion related to processing efficiency.

13. The system of claim 11 , wherein the entity graph is traversed to a depth based on a defined criterion related to a user input.

14. The system of claim 11 , wherein the collection component is further configured to:
transform a collection path into a canonical form; and
check an equivalent path to the canonical form to avoid redundant path traversal.

15. A computer program product to provide automatic feature extraction, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processing component to cause the processing component to:
generate an entity graph with a first node representing a first table of a relational database and a second node representing a second table of the relational database;
form a joined table based on an edge between the first node and the second node, wherein the joined table is connected by a column of data; and
extract, using a data mining algorithm, one or more features from tables of the relational database based on the column of data of the joined table.

16. The computer program product of claim 15 , wherein the program instructions are further executable by the processing component to cause the processing component to:
select the data mining algorithm from a set of data mining algorithms for the feature extraction based on determining a type of data in the column of data from a group consisting of spatial-temporal data, time-series data, sequence data, item set data, number set data, singleton data, text data, and image data.

17. The computer program product of claim 15 , wherein the program instructions are further executable by the processing component to cause the processing component to:
to select a feature from the one or more features based on a relevance to a target variable that is defined by an entity in a main table associated with a root node of the entity graph.

18. The computer program product of claim 15 , wherein the program instructions are further executable by the processing component to cause the processing component to:
collect the one or more features extracted from tables of the relational database by traversing the entity graph.

19. The computer program product of claim 18 , wherein the entity graph is traversed to a depth based on a defined criterion related to processing efficiency.

20. The computer program product of claim 18 , wherein the entity graph is traversed to a depth based on a defined criterion related to a user input.

Assignees

• IBM

Inventors

• Chen Bei
• Hoang Thanh Lam
• Mai Tiep
• Sinn Mathieu
• Thiebaut Johann-Michael

Classifications

• G06F2216/03

  Data mining · CPC title

• G06F16/2465

  Query processing support for facilitating data mining operations in structured databases · CPC title

• G06F16/9024

  Graphs; Linked lists (G06F16/9027 takes precedence) · CPC title

• G06F16/288 Primary

  Entity relationship models · CPC title