Ensemble-of-under-sampled extreme learning machine

Abstract

The invention relates to an ensemble-of-under-sampled extreme learning machine which is characterized in that for a training sample with class data imbalance, performing random under-sampling on a majority sample (FP data) in the training sample at first and then segmenting the majority sample into N majority subsamples according to a ratio N of the majority sample to a minority sample; combining the N majority subsamples with the minority sample respectively to form N training subsets; training N extreme learning machines by the obtained N training subsets to obtain N classifiers; feeding test samples to the N classifiers respectively, wherein each classifier obtains a classification result; setting a decision threshold value D, combining the classification results, and comparing a combined classification result with the decision threshold value D to decide a final classification result, wherein all the classifiers are same in voting weight. The ensemble-of-under-sampled extreme learning machine is relatively high in classification efficiency and simple in parameter adjustment method.

Description

Technical field [0001] The invention relates to pattern recognition technology, in particular to a classifier for unbalanced data sets. Background technique [0002] With the rapid development of information intelligence technology, machine learning technology is booming, its application fields are more extensive, and technology development is more in-depth. Classification is one of the important means of knowledge acquisition in machine learning and data mining. At present, many classification algorithms such as support vector machine (SVM) and neural network have been widely used in various fields. However, existing classification algorithms usually assume that the data set used for training is balanced, that is, the number of samples contained in each class is approximately equal. But most of the actual situation is not the case, especially in the field of medicine, in clinical cases, positive cases (that is, sick cases) are far less than negative cases (that is, normal ...

AI Technical Summary

Problems solved by technology

Although EUS-SVM overcomes the sample selection dependence and instability of simple downsampling, and can achieve a better minority sample detection rate on unbalanced samples, but as a base classifier, SVM takes a long time to train and the parameter The setting is more complicated. When the amount of training data is large, or the proportion of minority and majority samples is quite different, and multiple base classifiers need to be assembled, EUS-SVM needs to train multiple SVM classifiers, and its classification efficiency is more worrying.

Images