Authors: Zhi-ying Xu; YiJiang Zhang
Addresses: Shaoxing University Yuanpei College, Shaoxing 312000,China ' Shaoxing University Yuanpei College, Shaoxing 312000,China
Abstract: The classification of imbalanced data can increase the comprehensibility and expansibility of data and improve the efficiency of data classification. The accuracy of classification is poor when the data is classified by the current method for imbalanced data analysis of big data. To this end, this paper presents an imbalanced data classification algorithm based on fuzzy rule. The algorithm firstly collects the imbalanced data, selects the features of the imbalanced data, and optimises the imbalanced data classification algorithm by using the fuzzy rule classification algorithm. The experimental results show that when the classifier maintains a certain size of the weak classifier, the classification accuracy of the proposed algorithm will be gradually improved as the training time increases, and gradually be stable within a certain range of accuracy, this method can improve the accuracy of imbalanced data classification.
Keywords: imbalanced data; data feature selection; data classification.
International Journal of Information and Communication Technology, 2019 Vol.14 No.3, pp.373 - 384
Received: 11 Aug 2017
Accepted: 08 Sep 2017
Published online: 29 Mar 2019 *