Compact representation and time series segment retrieval through deep learning

Patent metadata

Field	Value
Publication number	US-2022012538-A1
Application number	US-202117364125-A
Country	US
Kind code	A1
Filing date	Jun 30, 2021
Priority date	Jul 7, 2020
Publication date	Jan 13, 2022
Grant date	—

Abstract

Systems and methods for retrieving similar multivariate time series segments are provided. The systems and methods include extracting a long feature vector and a short feature vector from a time series segment, converting the long feature vector into a long binary code, and converting the short feature vector into a short binary code. The systems and methods further include obtaining a subset of long binary codes from a binary dictionary storing dictionary long codes based on the short binary codes, and calculating similarity measure for each pair of the long feature vector with each dictionary long code. The systems and methods further include identifying a predetermined number of dictionary long codes having the similarity measures indicting a closest relationship between the long binary codes and dictionary long codes, and retrieving a predetermined number of time series segments associated with the predetermined number of dictionary long codes.

First claim

What is claimed is:

1 . A computer implemented method of retrieving similar multivariate time series segments, comprising:
extracting a long feature vector and a short feature vector from a time series segment; converting the long feature vector into a long binary code; converting the short feature vector into a short binary code; obtaining a subset of long binary codes from a binary dictionary storing dictionary long codes based on the short binary codes; calculating similarity measure for each pair of the long feature vector with each dictionary long code; identifying a predetermined number of dictionary long codes having the similarity measures indicting a closest relationship between the long binary codes and dictionary long codes; and retrieving a predetermined number of time series segments associated with the predetermined number of dictionary long codes.

2 . The computer implemented method as recited in claim 1 , further comprising displaying the predetermined number of time series segments to a user.

3 . The computer implemented method as recited in claim 1 , wherein the long feature vector and the short feature vector are extracted from the time series segments using a long short term memory (LSTM).

4 . The computer implemented method as recited in claim 3 , wherein the long feature vector is converted into a long binary code by checking the signs of all entries in the feature vector.

5 . The computer implemented method as recited in claim 4 , wherein the short feature vector is converted into a short binary code by a linear mapping.

6 . The computer implemented method as recited in claim 5 , further comprising classifying the short binary codes to a class.

7 . The computer implemented method as recited in claim 6 , wherein classifying involves computing the probability of the short binary code belong to each of a plurality of labels associated with the time series segments.

8 . A processing system for retrieving similar multivariate time series segments, comprising:
one or more processors; memory coupled to the one or more processors; a long feature extractor stored in memory, wherein the long feature extractor is configured to extract a long feature vector from a time series segment; a short feature extractor stored in memory, wherein the short feature extractor is configured to convert a long feature generated by the long feature extractor into a shorter length feature through a linear mapping; a long binary extractor stored in memory, wherein the long binary extractor is configured to convert a long feature from the long feature extractor into a long binary code having the same length as the long feature; a short binary extractor stored in memory, wherein the short binary extractor is configured to convert a short feature from the short feature extractor into a short binary code having the same length as the short feature; and a similarity comparator stored in memory, wherein the similarity comparator is configured to calculate a pairwise similarity between a long binary code extracted from the query and all long binary codes retrieved from a dictionary, and identifying a predetermined number of dictionary long codes having the similarity measures indicting a closest relationship between the long binary codes and dictionary long codes.

9 . The processing system as recited in claim 8 , wherein the short feature from the short feature extractor into a short binary code having the same length as the short feature by checking the sign of the entries in a short feature vector.

10 . The processing system as recited in claim 8 , wherein the similarity comparator is configured to retrieve a predetermined number of time series segments associated with the predetermined number of dictionary long codes, and display the predetermined number of time series segments to a user.

11 . The processing system as recited as recited in claim 10 , wherein the long feature vector and the short feature vector are extracted from the time series segments using a long short term memory (LSTM).

12 . The processing system as recited as recited in claim 11 , wherein the long feature vector is converted into a long binary code by checking the signs of all entries in the feature vector.

13 . The processing system as recited as recited in claim 12 , wherein the short feature vector is converted into a short binary code by a linear mapping.

14 . The processing system as recited as recited in claim 13 , wherein the short binary extractor is further configured to classifying the short binary codes to a class.

15 . A computer program product for retrieving similar multivariate time series segments, the computer program product comprising a non-transitory computer readable storage medium having program instructions embodied therewith, the program instructions executable by a computer to cause the computer to perform a method comprising:
extracting a long feature vector and a short feature vector from a time series segment; converting the long feature vector into a long binary code; converting the short feature vector into a short binary code; obtaining a subset of long binary codes from a binary dictionary storing dictionary long codes based on the short binary codes; calculating similarity measure for each pair of the long feature vector with each dictionary long code; identifying a predetermined number of dictionary long codes having the similarity measures indicting a closest relationship between the long binary codes and dictionary long codes; and retrieving a predetermined number of time series segments associated with the predetermined number of dictionary long codes.

16 . The computer program product as recited in claim 15 , further comprising displaying the predetermined number of time series segments to a user.

17 . The computer program product as recited in claim 15 , wherein the long feature vector and the short feature vector are extracted from the time series segments using a long short term memory (LSTM).

18 . The computer program product as recited in claim 17 , wherein the long feature vector is converted into a long binary code by checking the signs of all entries in the feature vector.

19 . The computer program product as recited in claim 18 , wherein the short feature vector is converted into a short binary code by a linear mapping.

20 . The computer program product as recited in claim 19 , further comprising classifying the short binary codes to a class, wherein classifying involves computing the probability of the short binary code belong to each of a plurality of labels associated with the time series segments.

Assignees

• Nec Lab America Inc

Inventors

• Mizoguchi Takehiko
• Song Dongjin
• Chen Yuncong
• Lumezanu Cristian
• Chen Haifeng

Classifications

• G06N3/084 Primary

  Backpropagation, e.g. using gradient descent · CPC title

• G06F18/2163

  Partitioning the feature space · CPC title

• G06F18/2415

  based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate · CPC title

• G06N3/044

  Recurrent networks, e.g. Hopfield networks · CPC title

• G06N3/045

  Combinations of networks · CPC title