Time sequence classification method, system, medium and device based on multi-representation learning

Abstract

The invention relates to a time sequence classification method, system, medium and device based on multi-representation learning. The method comprises the following steps: (1) carrying out the multi-feature coding of a given time sequence based on different time sequence representation strategies; (2) realizing representation fusion and enhancement by using a residual network and a bidirectional long-short-term memory network; and (3) completing classification by using a multi-layer perceptron network, and realizing classification interpretability by using an attention mechanism. According to the method, a multi-channel time sequence representation learning model is constructed, so that the time sequence characteristics can be comprehensively understood based on various representation strategies. According to the representation fusion model based on the residual network and the bidirectional long and short term memory network, multi-view representation can be effectively fused and representation enhancement can be realized, so that the classification precision is effectively improved. According to the method, the important time sequence characteristics of the time sequence can be effectively recognized on the basis of the attention mechanism, namely, the interpretability basis of the classification result can be provided, and the classification interpretability is realized.

Description

technical field [0001] The invention belongs to the technical field of time series data mining, and in particular relates to a time series classification method, system, medium and equipment based on multi-representation learning. Background technique [0002] With the rapid development of technologies such as the Internet, cloud computing, and big data processing, we have entered the era of big data where everything is connected. Among the massive heterogeneous data, there is a kind of time-related time series data, which is widely used in almost all application fields in the real world, and is called "time series" for short. Time series can not only reflect the specific data characteristics at a certain moment, but also reveal the continuous change over time, the change trend and potential knowledge of data entities. Time series often have big data characteristics such as "massive", "high-dimensional", and "continuously generated", and it is very challenging to study them...

AI Technical Summary

Problems solved by technology

The limitations of the existing TSC methods are: on the one hand, the timing feature extraction strategy is combined with a variety of traditional classification algorithms, and then the voting mechanism is used to improve the classification accuracy to a certain extent, while ignoring the potential contribution of deep learning to representation learning; On the other hand, the existing TSC method can only simply use a single deep learning model (convolutional neural network, bidirectional long-short-term memory network) to learn general representations of time series and complete classification, lacking the adaptive understanding and understanding of various time series features. representation, so it may not be able to effectively improve the classification accuracy

Therefore, it is equally challenging to provide an interpretable basis for classification results

Images