Unbalanced time series data classification method based on autonomous learning

Abstract

The invention discloses an unbalanced time series data classification method based on autonomous learning, and the method specifically comprises the following steps: 1, carrying out the processing of unbalanced time series data, and constructing a new sample; 2, sequentially performing scale transformation and data segmentation on the new sample constructed in the step 1; 3, constructing a deep convolutional neural network model based on a result obtained in the step 2; and 4, training the neural network model constructed in the step 3, and establishing an optimal time sequence data classification model according to a training result to perform time sequence classification. According to the method, the problem that the minority class detection precision is seriously reduced due to the fact that a general learner is absolutely biased to the majority class is solved, and the classification precision of the unbalanced time series data set is remarkably improved.

Description

technical field [0001] The invention belongs to the technical field of time series data classification, and relates to an unbalanced time series data classification method based on autonomous learning. Background technique [0002] Time series refers to data arranged in chronological order, which can directly reflect the state or degree of a certain thing or phenomenon over time; time series data mining is to extract from a large amount of time series data that people do not know in advance, Useful information related to time attributes is used to guide people's social, economic, and life activities. In the field of aerospace measurement and control, a large amount of telemetry data is presented in the form of time series. These engineering data can directly reflect the operating status of the aircraft. It is very important to classify these data and dig out the information and laws contained in them for the research of equipment fault diagnosis technology. of. Therefore, ...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a classification method for unbalanced time series data based on autonomous learning, which solves the problem that the general learner is absolutely biased towards the majority class, thereby causing a serious decline in the detection accuracy of the minority class, and significantly improves the unbalanced time series data classification method. Classification Accuracy of the Dataset

Images