An Incremental Learning of Concept Drifts Using Evolving Type-2 Recurrent Fuzzy Neural Networks

Mahardhika Pratama, Jie Lu, Edwin Lughofer, Guangquan Zhang, Meng Joo Er
2017 IEEE transactions on fuzzy systems  
the age of big data and dynamic environments result in the increasing demand of advanced machine learning techniques to deal with concept drifts in large data streams. Evolving Fuzzy Systems (EFS) are one of recent initiatives from the fuzzy system community to resolve the issue. Existing EFSs are not robust against data uncertainty, temporal system dynamics, and the absence of system order, because vast majority of EFSs are designed in the feed-forward type-1 fuzzy network architecture. This
more » ... per aims to solve the issue of data uncertainty, temporal behaviour, and the absence of system order by developing a novel evolving recurrent fuzzy neural network, called Evolving Type-2 Recurrent Fuzzy Neural Network (eT2RFNN). eT2RFNN is constructed in a new recurrent network architecture, featuring double recurrent layers. The new recurrent network architecture evolves a generalized interval type-2 fuzzy rule, where the rule premise is built upon the interval type-2 multivariate Gaussian function, while the rule consequent is crafted by the non-linear wavelet function. The eT2RFNN adopts a holistic concept of evolving systems, where the fuzzy rule can be automatically generated, pruned, merged and recalled in the single pass learning mode. eT2RFNN is capable of coping with the problem of high dimensionality, because it is equipped with online feature selection technology. The efficacy of eT2RFNN has been experimentally validated using artificial and real-world data streams and compared with prominent learning algorithms. eT2RFNN can produce more reliable predictive accuracy, while retaining lower complexity than its counterparts. Index Terms-evolving fuzzy systems, fuzzy neural networks, recurrent fuzzy neural networks, type-2 fuzzy systems, incremental learning, concept drifts I.
doi:10.1109/tfuzz.2016.2599855 fatcat:6pvyhj22e5dsviht6aznjdqdzq