Construction of reference database accurately representing complete set of data items for faster and tractable classification usage

Patent metadata

Field	Value
Publication number	US-11314781-B2
Application number	US-201816146576-A
Country	US
Kind code	B2
Filing date	Sep 28, 2018
Priority date	Sep 28, 2018
Publication date	Apr 26, 2022
Grant date	Apr 26, 2022

Abstract

For each unique pair of a complete set of data items, a computing device determines a distance between the data items of the unique pair. The computing device repeats the following until no data items remain in the complete set. For each data item remaining in the complete set, the computing device determines a similarity subset including each other data item that the distance between the data item and the other data item is less than a target difference threshold. The computing device moves a selected data item from a largest similarity subset to a reference database that is a subset of the complete set. The computing device removes each data item from the complete set that the distance between the selected data item and the data item is less than the threshold. A new data item can be classified using the reference database.

First claim

We claim:

1. A method comprising:
storing, by a computing device on a storage device, a reference database that is a subset of a complete set of data items, the reference database requires less storage space than the complete set of data items, each data item of the complete set having a distance to a data item of the subset that is less than a target difference threshold, the reference database having a classification accuracy equal to the k-th power of one minus the target difference threshold, and wherein k is an associated length of each data item; and
classifying, by the computing device using logic implemented at least in hardware, a new data item using the reference database.

2. The method of claim 1 , wherein each data item is a genome having a plurality of substrings that are each of length k.

3. The method of claim 2 , wherein each the plurality of substrings is a kmer.

4. The method of claim 1 , wherein the classification accuracy is a probability that a new data item is classified correctly using just the data items moved to the reference database and not all the data items of the complete set.

5. The method of claim 1 , wherein classifying the new data item using just the reference database and not the complete set occurs more quickly than using the complete set.

6. The method of claim 1 , wherein the target difference threshold is represented as a percentage.

7. The method of claim 1 , wherein each data item is a protein having one or more sub strings that are each of length k.

8. A computer system comprising:
one or more computer processors, one or more computer-readable storage media, and program instructions stored on the one or more computer-readable storage media for execution by at least one of the one or more computer processors, the program instructions comprising:
program instructions to store, by a computing device on a storage device, a reference database that is a subset of a complete set of data items, the reference database requires less storage space than the complete set of data items, each data item of the complete set having a distance to a data item of the subset that is less than a target difference threshold, wherein the reference database has a classification accuracy equal to the k-th power of one minus the target difference threshold, and wherein k is an associated length of each data item; and
program instructions to classify, by the computing device using logic implemented at least in hardware, a new data item using the reference database.

9. The computer system of claim 8 , wherein each data item is a genome having a plurality of sub strings that are each of length k.

10. The computer system of claim 9 , wherein each of the plurality of substrings is a kmer.

11. The computer system of claim 8 , wherein the classification accuracy is a probability that a new data item is classified correctly using just the data items moved to the reference database and not all the data items of the complete set.

12. The computer system of claim 8 , wherein classifying the new data item using just the reference database and not the complete set occurs more quickly than using the complete set.

13. The computer system of claim 8 , wherein the target difference threshold is represented as a percentage.

14. The computer system of claim 8 , wherein each data item is a protein having one or more substrings that are each of length k.

15. A computer program product comprising:
one or more computer-readable storage media and program instructions stored on the one or more computer-readable storage media, the program instructions comprising:
program instructions to store, by a computing device on a storage device, a reference database that is a subset of a complete set of data items, the reference database requires less storage space than the complete set of data items, each data item of the complete set having a distance to a data item of the subset that is less than a target difference threshold, wherein the reference database has a classification accuracy equal to the k-th power of one minus the target difference threshold, and wherein k is an associated length of each data item; and
program instructions to classify, by the computing device using logic implemented at least in hardware, a new data item using the reference database.

16. The computer system of claim 15 , wherein each data item is a genome having a plurality of substrings that are each of length k.

17. The computer system of claim 16 , wherein each the plurality of substrings is a kmer.

18. The computer system of claim 15 , wherein the classification accuracy is a probability that a new data item is classified correctly using just the data items moved to the reference database and not all the data items of the complete set.

19. The computer system of claim 15 , wherein classifying the new data item using just the reference database and not the complete set occurs more quickly than using the complete set.

20. The computer system of claim 15 , wherein the target difference threshold is represented as a percentage.

21. The computer system of claim 15 , wherein each data item is a protein having one or more sub strings that are each of length k.

Assignees

• IBM
• Mars Inc

Inventors

• Kaufman James H
• Davis Matthew A
• Weimer Bart C

Classifications

• G06F16/285 Primary

  Clustering or classification · CPC title