Online Evolving Interval Type-2 Intuitionistic Fuzzy LSTM-Neural Networks for Regression Problems

Abstract

Since the existence of fuzziness in the real world, uncertainty could be always present in the reasoning results. Therefore, how to tackle with the hesitation existing in the process of reasoning is a meaningful issue for the construction of the inference system. In this paper, a novel interval type-2 intuitionistic fuzzy neural network based on long-short term mechanism is proposed (LSTM-IT2IFNN). By means of interval type-2 intuitionistic set, the hesitation of reasoning is described and involved to determine fuzzy rules in inference system, by which the chosen of fuzzy rules not only considers the membership value but also non-membership degree. In order to handle the regression problems with long-term time dependency, long-short term mechanism is introduced into fuzzy neural networks, which can effectively improve the memory ability of long-term knowledge. Furthermore, by means of intuitionistic sets, an extended metacognitive learning mechanism is proposed for the adaptively online learning of network structure, where it achieves when, what, and how to learn from the sequential data. The performance of the LSTM-IT2IFNN is evaluated on different regression problems, including synthetic and real-world data. By comparing to existing algorithms, the experimental results show the efficiency and accuracy of the proposed scheme. INDEX TERMS Fuzzy neural network, intuitionistic fuzzy sets, interval type-2 fuzzy systems, metacognition, sequential learning.

Untitled

“…Experimental results reveal that EKF-based learning models perform better than the GD-based in terms of prediction accuracy. In [11], a hybrid model of GD back-propagation and DEKF is employed for the adjustment of the parameters of IT2IFLS and the model applied to system identification problem.Recently also, Yuan and Luo [12] proposed an online evolving interval type-2 intuitionistic fuzzy LSTM-Neural Networks (eIT2IF-LSTMNN). In [12] and [13], the parameters of the models are optimized using GD-back-propagationand applied for regression problems.Other studies on IT2IFS is the one reported in [14], where arithmetic operations are defined for IT2IFS using generalized trapezoidal type-2 intuitionistic fuzzy numbers.…”

“…The test signal in (34) is adopted to test the quality of prediction. Similar to [12], we adopt the same training data and time steps as in system identification 1. To reduce the complexity of the system, two inputs u(t) andy(t)are passed into the IT2IFLS.…”

Elliptic interval Type-2 intuitionistic fuzzy logic system for non-linear system identification

“…These present IT2IFS as a different architecture from IVIFS. The interval type-2 intuitionistic fuzzy logic systems (IT2IFLSs) have found applications in many problem domains such as time series (Eyoh et al, 2017, Luo et al, 2019, regression problems (Eyoh et al, 2018a, Yuan andChao, 2019) and identification and prediction problems (Eyoh et al, 2018b).…”

Software Fault Prediction Based on Interval Type-2 Intuitionistic Fuzzy Logic System

“…However, because the IFS of type-1 is designed around a single point of view and may not capture various forms of uncertainties, Eyoh et al [31] presented a rule-based interval type-2 intuitionistic fuzzy logic system (IT2IFLS) that are defined using fuzzy MFs and NMFs together with hesitation indexes. The IT2IFLS has successfully been applied to solve many applications in different domains with impressive performance such as load forecasting [35], time series prediction [36], [37], [38], [39], [40], [41], identification and prediction problems [31], [42], [43], regression problems [44], [45] transportation problems [46] and many more. To the best knowledge of the authors, this is the first study that adopts IT2IFLS with MF, NMF and hesitation indices for software fault prediction.…”