Facilitating data-driven mapping discovery

Patent metadata

Field	Value
Publication number	US-2019130309-A1
Application number	US-201715798493-A
Country	US
Kind code	A1
Filing date	Oct 31, 2017
Priority date	Oct 31, 2017
Publication date	May 2, 2019
Grant date	—

Abstract

Systems, computer-implemented methods and/or computer program products that facilitate automatically mapping different data types are provided. In one embodiment, a computer-implemented method comprises: constructing, by a system operatively coupled to a processor, an index from one or more classifier models for one or more data types; scoring and ranking, by the system, one or more candidate pairs for the one or more data types based on confidence score; and analyzing, by the system, how the one or more candidate pairs are scored and automatically generating the one or more classifier models used to construct the index.

First claim

What is claimed is:

1 . A system, comprising:
a memory that stores computer executable components; a processor, operably coupled to the memory, and that executes computer executable components stored in the memory, wherein the computer executable components comprise:
an index component that constructs an index from one or more classifier models for one or more data types;
one or more scoring components that score and rank one or more candidate pairs for the one or more data types based on confidence score; and
a machine learning component that analyzes how the one or more candidate pairs are scored and automatically generates the one or more classifier models used to construct the index.

2 . The system of claim 1 , wherein the machine learning component also collects data used to generate the one or more classifier models.

3 . The system of claim 1 , wherein the machine learning component also automatically generates one or more maps used to automatically generate the one or more classifier models.

4 . The system of claim 1 , further comprising an output component that produces priority levels for the one or more candidate pairs based on a determination that the confidence score is equal to or greater than a defined threshold.

5 . The system of claim 4 , wherein the machine learning component also constructs a new classifier model if the confidence score of the one or more candidate pairs for the one or more data types is below the defined threshold.

6 . The system of claim 1 , further comprising a search component that searches the index for the one or more candidate pairs from the one or more data types.

7 . The system of claim 6 , wherein the machine learning component selects the one or more candidate pairs to train the one or more classifier models based on an analysis of how the one or more candidate pairs are scored by comparing different confidence scores from the one or more classifier models of the one or more data types.

8 . The system of claim 7 , wherein the one or more scoring components modify one or more scoring parameters and the defined threshold if the one or more candidate pairs selected to train the one or more classifier models are few.

9 . A computer-implemented method, comprising:
constructing, by a system operatively coupled to a processor, an index from one or more classifier models for one or more data types; scoring and ranking, by the system, one or more candidate pairs for the one or more data types based on confidence score; and analyzing, by the system, how the one or more candidate pairs are scored and automatically generating the one or more classifier models used to construct the index.

10 . The computer-implemented method of claim 9 , further comprising using the machine learning component to automatically generate one or more maps used to automatically generate the one or more classifier models.

11 . The computer-implemented method of claim 9 , further comprising using an output component to produce priority levels for the one or more candidate pairs based on a determination that the confidence score is equal to or greater than a defined threshold.

12 . The computer-implemented method of claim 11 , further comprising using the machine learning component to construct a new classifier model if the confidence score of the one or more candidate pairs for the one or more data types is below the defined threshold.

13 . The computer-implemented method of claim 9 , further comprising using a search component to search the index for the one or more candidate pairs from the one or more data types.

14 . The computer-implemented method of claim 13 , further comprising using the machine learning component to select the one or more candidate pairs to train the one or more classifier models based on an analysis of how the one or more candidate pairs are scored by comparing different confidence scores from the one or more classifier models of the one or more data types.

15 . A computer program product for facilitating automatically mapping different data types, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to:
construct an index from one or more classifier models for one or more data type; score and rank one or more candidate pairs for the one or more data types based on confidence score; and analyze how the one or more candidate pairs are scored and automatically generate the one or more classifier models used to construct the index.

16 . The computer program product of claim 15 , wherein the program instructions are further executable to cause the processor to:
automatically generate one or more maps used to automatically generate the one or more classifier models.

17 . The computer program product of claim 15 , wherein the program instructions are further executable to cause the processor to:
produce priority levels for the one or more candidate pairs based on a determination that the confidence score is equal to or greater than a defined threshold.

18 . The computer program product of claim 17 , wherein the program instructions are further executable to cause the processor to:
construct a new classifier model if the confidence score of the one or more candidate pairs for the one or more data types is below the defined threshold.

19 . The computer program product of claim 15 , wherein the program instructions are further executable to cause the processor to:
search the index for the one or more candidate pairs from the one or more data types.

20 . The computer program product of claim 19 , wherein the program instructions are further executable to cause the processor to:
select the one or more candidate pairs to train the one or more classifier models based on an analysis of how the one or more candidate pairs are scored by comparing different confidence scores from the one or more classifier models of the one or more data types.

Assignees

• IBM

Inventors

• Jiang Jian Min
• Liu Pei Ni
• Ni Yuan
• Sun Wen
• Xie Guo Tong
• Xu Jing Min

Classifications

• G06N20/10

  using kernel methods, e.g. support vector machines [SVM] · CPC title

• G06N5/022

  Knowledge engineering; Knowledge acquisition · CPC title

• G06F16/901 Primary

  Indexing; Data structures therefor; Storage structures (for retrieval from the web G06F16/951) · CPC title

• G06F16/906

  Clustering; Classification · CPC title

• G06N20/00 Primary

  Machine learning · CPC title