Wavelet transform-based fine-granularity self-learning integration prediction method

Abstract

The invention discloses a wavelet transform-based fine-granularity self-learning integration prediction method which comprises the following steps: by adopting time sequence decomposition based on wavelet transform, predicting time sequences of different variable coefficients with different granularities, so as to relatively precisely reveal the variation rules of the time sequences; with the combination of the time sequence decomposition, extracting characteristics of a plurality of related factors, sufficiently capturing main influence factors, and predicting future trend through rule statistics, thereby being rapid, convenient and simple; applying a model-based aggregation algorithm frame to the regression process, thereby enabling the model to have robustness which is relatively good when being compared with that of a single based learner; predicting with the combination of composite models based on wavelet transform, SVR and Ensemble, thereby obtaining prediction performance which is relatively precise when being compared with that of a conventional single model. The wavelet transform-based fine-granularity self-learning integration prediction method can be widely applied to the technical field of big-data mining and machine learning.

Description

technical field [0001] The invention relates to the technical field of big data mining and machine learning, and more particularly to a fine-grained self-learning integrated prediction method based on wavelet transform. Background technique [0002] Common data mining algorithms for prediction, such as linear regression, regression tree, neural network, etc., have been adopted by most researches and applied in different fields. However, these methods simply input all the features into a single model for prediction, and do not consider the temporal features, the ability to capture the trend and peak value of the time series with high coefficient of variation is poor, and the generalization ability is weak. lead to low prediction accuracy. [0003] Therefore, it is necessary to provide a compound prediction model for more accurate data mining and prediction, which can be based on output features, input features, time features, lag features, granularity features, spatial featu...

AI Technical Summary

Problems solved by technology

However, these methods simply input all features into a single model for prediction, and do not consider time features. They have poor ability to capture the trend and peak value of time series with high coefficient of variation, and weak generalization ability. lead to low prediction accuracy

Images