# US 2020/0004616 A1 · Failure prediction

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2020/0004616 A1
- Document ID: US20200004616A1
- Status: Published application
- Kind: A1
- Application number: 16/458,687
- Filed: 2019-07-01
- Published: 2020-01-02
- Earliest claimed: 2018-07-02
- Assignees: NEC Laboratories America, Inc.
- Applicants: NEC Laboratories America, Inc.
- Inventors: Masanao Natsumeda; Wei Cheng; Haifeng Chen
- Web page: https://patentgrep.com/patent/US20200004616A1

## CPC

- G06F 11/008: {Reliability or availability analysis}
- G06N 3/08: Learning methods
- G06N 3/044: Recurrent networks, e.g. Hopfield networks
- G06N 3/0442: characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
- G06N 3/045: Combinations of networks
- G06N 3/09: Supervised learning
- G06N 5/02: Knowledge representation; Symbolic representation
- G06F 11/302: {where the computing system component is a software system}
- G06F 11/3447: {Performance evaluation by modeling}
- G06F 11/3452: {Performance evaluation by statistical analysis}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

A failure prediction system is provided. The system includes a model-based signature generator generating feature vectors from individual attributes of multi-variate time series data based on sequence importance and attribute importance. The system further includes a knowledge database storing feature vectors corresponding to a set of different failure types. The system also includes a set of similarity detectors. Each detect any of the feature vectors generated by the model-based signature generator that are similar to any of the feature vectors corresponding to a respective one of the different failure types stored in the knowledge database based on a similarity threshold and output the respective one of the different failure types and a likely time period when the respective one of the different failure types will occur.

## Claims

### Claim 1

A failure prediction system, comprising:
a model-based signature generator generating feature vectors from individual attributes of multi-variate time series data based on sequence importance and attribute importance;
a knowledge database storing feature vectors corresponding to a set of different failure types; and
a set of similarity detectors, each detecting any of the feature vectors generated by the model-based signature generator that are similar to any of the feature vectors corresponding to a respective one of the different failure types stored in the knowledge database based on a similarity threshold and outputting the respective one of the different failure types and a likely time period when the respective one of the different failure types will occur.

### Claim 2 (depends on 1)

The failure prediction system of claim 1, wherein the signal generator generates the feature vectors as a matrix.

### Claim 3 (depends on 1)

The failure prediction system of claim 1, wherein the signal generator comprises:
a feature extractor extracting feature values from the individual attributes of the multi-variate time series and concatenating the feature values into the feature vectors;
a feature selector assigning weights to each of the feature vectors; and
a feature converter converting multiple weighted ones of the feature vectors into a new feature vector.

### Claim 4 (depends on 3)

The failure prediction system of claim 3, wherein the feature extractor comprises a set of Recurrent Neural Networks.

### Claim 5 (depends on 3)

The failure prediction system of claim 3, wherein the feature extractor, feature selector, and feature converter are jointly trained.

### Claim 6 (depends on 1)

The failure prediction system of claim 1, wherein the feature vectors corresponding to the set of different failure types form a historical failure record of a target system sourcing the multi-variate time series data.

### Claim 7 (depends on 1)

The failure prediction system of claim 1, wherein the feature vectors corresponding to the set of different failure types are used to form a prediction model.

### Claim 8 (depends on 1)

The failure prediction system of claim 1, wherein the model is trained by minimizing a triplet loss involving a first distance between an anchor and a positive sample, a second distance between the anchor and a negative sample, and a margin between the first and second distances.

### Claim 9 (depends on 1)

The failure prediction system of claim 1, wherein the multi-variate time series data is divided into a set of different regions, the set of different regions comprising normal regions and early symptom regions.

### Claim 10

A computer-implemented method for failure prediction, comprising:
generating, by a model-based signature generator, feature vectors from multi-variate time series data based on sequence importance and attribute importance;
storing, by a knowledge database, feature vectors corresponding to a set of different failure types; and
detecting, by each of similarity detectors in a set, any of the feature vectors generated by the model-based signature generator that are similar to any of the feature vectors corresponding to the set of different failure types stored in the knowledge database based on a similarity threshold and outputting the respective one of the different failure types and a likely time period when the respective one of the different failure types will occur.

### Claim 11 (depends on 10)

The computer-implemented method of claim 10, wherein the feature vectors are generated by the signal generator as a matrix.

### Claim 12 (depends on 10)

The computer-implemented method of claim 10, wherein said generating step comprises:
extracting, by a feature extractor, feature values from the individual attributes of the multi-variate time series and concatenating the feature values into the feature vectors;
assigning, by a feature selector, weights to each of the feature vectors; and
converting, by a feature converter, multiple weighted ones of the feature vectors into a new feature vector.

### Claim 13 (depends on 12)

The computer-implemented method of claim 12, wherein the feature extractor comprises a set of Recurrent Neural Networks.

### Claim 14 (depends on 12)

The computer-implemented method of claim 12, further comprising jointly training the feature extractor, feature selector, and feature converter.

### Claim 15 (depends on 10)

The computer-implemented method of claim 10, wherein the feature vectors corresponding to the set of different failure types form a historical failure record of a target system sourcing the multi-variate time series data.

### Claim 16 (depends on 10)

The computer-implemented method of claim 10, further comprising forming a prediction model using the feature vectors corresponding to the set of different failure types.

### Claim 17 (depends on 10)

The computer-implemented method of claim 10, wherein the model is trained by minimizing a triplet loss involving a first distance between an anchor and a positive sample, a second distance between the anchor and a negative sample, and a margin between the first and second distances.

### Claim 18

A computer program product for failure prediction, 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:
generating, by a model-based signature generator of the computer, feature vectors from multi-variate time series data based on sequence importance and attribute importance;
storing, by a knowledge database of the computer, feature vectors corresponding to a set of different failure types; and
detecting, by each of similarity detectors of the computer, any of the feature vectors generated by the model-based signature generator that are similar to any of the feature vectors corresponding to the set of different failure types stored in the knowledge database based on a similarity threshold and outputting the respective one of the different failure types and a likely time period when the respective one of the different failure types will occur.

### Claim 19 (depends on 18)

The computer program product of claim 18, wherein the feature vectors are generated by the signal generator as a matrix.

### Claim 20 (depends on 18)

The computer program product of claim 18, wherein said generating step comprises:
extracting, by a feature extractor, feature values from the individual attributes of the multi-variate time series and concatenating the feature values into the feature vectors;
assigning, by a feature selector, weights to each of the feature vectors; and
converting, by a feature converter, multiple weighted ones of the feature vectors into a new feature vector.
