Hierarchical Bagging method for sentiment analysis based on electroencephalogram signals

Abstract

The invention belongs to the technical field of electroencephalogram signal processing. The invention discloses a hierarchical Bagging method for sentiment analysis based on electroencephalogram signals. The hierarchical Bagging method comprises the steps of electroencephalogram sample data preprocessing, feature extraction and feature selection, replacement sampling of a training set, training ofa plurality of data subsets through different base classification algorithms and voting of a plurality of classifiers to obtain a classification result. Different from a traditional Bagging algorithmin which a single training subset corresponds to a single classification algorithm, hierarchical Bagging enables a plurality of training subsets to correspond to a single classification algorithm, and the risk that a classification algorithm with good single performance is deleted due to the fact that the classification algorithm does not adapt to individual data is reduced. According to the method, the accuracy of electroencephalogram signal classification can be effectively improved, the problem of poor stability of a single classification algorithm is solved, and the method can also be popularized to other similar types of data processing. The method is of great significance to emotion monitoring, risk prediction and supervised learning classification.

Description

technical field [0001] The invention belongs to the technical field of EEG signal processing, and in particular relates to a hierarchical bagging method for emotional analysis based on EEG signals. Background technique [0002] In today's digital and computerized era, the analysis and research of EEG signals plays an important role in the field of processing human's advanced thinking activities. In previous studies, emotion classification was mainly measured by the subjects' expressions, voices, body movements and other indicators, but because these indicators are greatly affected by personal habits and expressions, and can be camouflaged and concealed, compared with Under the circumstance, the method of using the physiological signal of brain waves to classify emotions has gained unique advantages of authenticity and accuracy. [0003] At present, the most commonly used existing technology for sentiment analysis using EEG signals is to apply the formatted EEG data to exist...

AI Technical Summary

Problems solved by technology

[0014] The Bagging algorithm has excellent balance error characteristics, but each classifier is trained only once, and the sample set selected for this training is only a small part of the initial training set. In this case, it is prone to unsatisfactory training results caused by accidental errors or sampling errors.

That is: because part of the data is not suitable for a well-performing algorithm, the algorithm is eliminated in the final voting screening of the integration

[0015] To sum up, the problem existing in the existing technology is: due to the complex data structure of the EEG signal, the Bagging method of integrated learning can reduce the problem of low accuracy caused by insufficient learning and inadaptability, but the Bagging algorithm is not as good as the base classifier. When the number is small, due to the low sampling ratio, it is easy to cause the classification algorithm with good performance in the voting step to be eliminated due to the inability to adapt to individual data.

For this reason, when the Bagging algorithm is used for EEG signal classification, the accuracy of the algorithm can never exceed the best-performing base classifier, so that the current research on ensemble learning applied to EEG classification algorithms only improves stability. without taking into account the accuracy

[0016] The difficulty in solving the above technical problems lies in that a single classification algorithm does not have a rich learning perspective and cannot adapt to high-dimensional and complex EEG data, and multiple classification algorithms cannot guarantee that each classification algorithm performs well by using a single operation with return voting. When encountering data suitable for the algorithm during the random sampling process

Images