Data flow classification algorithm based on AAE-DWMAL-LearnNSE (Automatic Assisted Engineering-Discrete Wavelength Multiple Input Multiple Output-LearnNSE)

Abstract

The invention relates to a data flow classification algorithm based on AAE-DWMIL-LearnNSE, which is a method for classifying data flows, belongs to the field of data mining and machine learning, and is characterized by comprising the following steps of: (1) updating the weight of each classifier; (2) carrying out weighted processing on the base classifier; (3) weighting and synthesizing an integrated classifier; (4) calculating a classification error rate on the new data set; (5) creating a comprehensive prediction model; and (6) determining and classifying a data flow classification function.The problem that the classification precision is greatly influenced due to lack of selective selection of the training samples is solved, the link of storing old data samples is omitted, most of storage space is saved for new data, meanwhile, an old classification model is fully utilized, and high classification accuracy is obtained. The problem of similarity between labeled training samples in the measurement source field and unlabeled test samples in the target field is effectively solved.

Description

technical field [0001] The invention relates to the fields of data mining and machine learning, and mainly relates to a method for identifying and classifying data streams. Background technique [0002] At present, for the classification of data streams, some traditional algorithms or improved algorithms in data mining are mainly used. The general processing flow requires a large number of labeled samples for training the classification model. Furthermore, due to some inherent characteristics of data streams, it is easy to cause difficult sample labeling and frequent concept drift in traditional data stream classification methods. The traditional incremental and integrated data stream classification methods can achieve better results in some fields, but they also have the disadvantages of gradually reducing the classification effect as the amount of data increases and the parameter explosion caused by the sharp increase in the number of base classifiers. In addition, the d...

AI Technical Summary

Problems solved by technology

Algorithms such as IS3RS and IDS-ELM lack selective selection of training samples, which will greatly affect the classification accuracy

Algorithms such as SEA and CELM are much more accurate than the results predicted by a single model, but there are problems such as how to select, update, and assign base classifiers

Images