Uncertain fuzzy self-organization based clustering: interval type-2 fuzzy approach to adaptive resonance theory

Majeed, Shakaiba; Gupta, Aditya; Raj, Desh; Rhee, Frank Chung-Hoon

doi:10.1016/j.ins.2017.09.062

Cited by 13 publications

(12 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…af is the adjustment coefficient [18], which is used to adjust the distance threshold. A large number of experiments show that when A is set to 0.3, the proposed algorithm could obtain good sorting effect.…”

Section: Parameter Settingsmentioning

confidence: 99%

An improved de-interleaving algorithm of radar pulses based on SOFM with self-adaptive network topology

Jiang

Chang

et al. 2020

J. of Syst. Eng. Electron.

View full text Add to dashboard Cite

As a core part of the electronic warfare (EW) system, de-interleaving is used to separate interleaved radar signals. As interleaved radar pulses become more complex and denser, intelligent classification of radar signals has become very important. The self-organizing feature map (SOFM) is an excellent artificial neural network, which has huge advantages in intelligent classification of complex data. However, the de-interleaving process based on SOFM is faced with the problems that the initialization of the map size relies on prior information and the network topology cannot be adaptively adjusted. In this paper, an SOFM with self-adaptive network topology (SANT-SOFM) algorithm is proposed to solve the above problems. The SANT-SOFM algorithm first proposes an adaptive proliferation algorithm to adjust the map size, so that the initialization of the map size is no longer dependent on prior information but is gradually adjusted with the input data. Then, structural optimization algorithms are proposed to gradually optimize the topology of the SOFM network in the iterative process, constructing an optimal SANT. Finally, the optimized SOFM network is used for de-interleaving radar signals. Simulation results show that SANT-SOFM could get excellent performance in complex EW environments and the probability of getting the optimal map size is over 95% in the absence of priori information.

show abstract

Section: Parameter Settingsmentioning

confidence: 99%

An improved de-interleaving algorithm of radar pulses based on SOFM with self-adaptive network topology

Jiang

Chang

et al. 2020

J. of Syst. Eng. Electron.

View full text Add to dashboard Cite

show abstract

“…In general, the type-1 fuzzy set (T1 FS) has been widely used to represent pattern uncertainty in the field of pattern recognition. However, as previously shown, T1 FS cannot produce good result and be extended to type-2 fuzzy set (T2 FS) in order to control the uncertain fuzzifier value more efficiently [21] [22]. T2 FS, ̃ , is represented as follows.…”

Section: Interval Type-2 Fuzzy Membership Functionmentioning

confidence: 99%

“…In general, the kernel method is to convert the input data from the input property space to the kernel property space through the kernel function using a space conversion function [22]. This is to change the kernel property space into the kernel property space making it easier to distinguish data that has or overlaps a non-linear boundary surface of input property space through kernel property space conversion.…”

Section: A Multiple Kernels Pfcm Algorithmmentioning

confidence: 99%

Data Clustering Method Using Efficient Fuzzifier Values Derivation

Cho

Joo

2020

IEEE Access

View full text Add to dashboard Cite

The Type-2 fuzzy set (T2 FS) is widely used for efficient control uncertainties, such as noise sensitivity in the fuzzy set. In addition, unsupervised machine learning requires a clustering parameter value in advance, and may affect clustering performance according to prior information such as the number and size of clusters. In this case, the fuzzifier value m to be applied is the most important factor in improving the accuracy of data. Therefore, in this paper, we intend to perform clustering to automatically acquire the determination of m1 and m2 values that depended on existing repeated experiments. To this end, in order to increase efficiency on deriving appropriate fuzzifier value, we used the Interval type-2 possibilistic fuzzy Cmeans (IT2PFCM), clustering method to classify a given pattern. In Efficient IT2PFCM method, used for clustering, we propose an algorithm that derives suitable fuzzifier values for each data. These values also extract information from each data point through the histogram approach and Gaussian Curve Fitting method. Using the extracted information, two adaptive fuzzifier value m1 and m2 are determined. Obtained values apply the new lowest and highest membership values. In addition, it is possible to form an appropriate fuzzy area on each cluster by only taking advantage of the characteristics of IT2PFCM, which reduces uncertainty. This doesn't only improve the accuracy of clustering of measured sensor data, but can also be used without additional procedures such as data labeling or the provision of prior information. It is also efficient at monitoring numerous sensors, managing and verifying sensor data collected in real time such as smart cities. Eventually, in this study, the proposed method is to improve IT2PFCM performance on accurate and quick clustering of large amount of complex data such as Internet of Things (IoT).

show abstract

“…The same running time complexity analysis applies to other ART neural architectures that faithfully follow the same learning algorithm. A thorough discussion of fuzzy ART computational complexity analysis was presented in (Granger et al, 1998), and summarized in other studies such as (Majeed et al, 2018;Meng et al, 2016Meng et al, , 2014.…”

Section: Speedmentioning

confidence: 99%

“…However, it is often set empirically in an ad hoc manner. In unsupervised learning mode, vigilance adaptation has been addressed in fuzzy ART through the activation maximization, confliction minimization and hybrid integration rules (Meng et al, 2013(Meng et al, , 2016; the combination with particle swarm optimization (Kennedy & Eberhart, 1995) and cluster validity indices (Xu & Wunsch II, 2009) in (Smith & Wunsch II, 2015); defining the vigilance as a function of the category size (Isawa et al, 2008b(Isawa et al, , 2009); or modeling it as a fuzzy membership function (Majeed et al, 2018). Despite these contributions, setting the vigilance parameter still remains a challenging task worthy of further exploration, particularly in the online learning mode.…”

Section: Vigilance Parameter Adaptationmentioning

confidence: 99%

A survey of adaptive resonance theory neural network models for engineering applications

2019

View full text Add to dashboard Cite

This survey samples from the ever-growing family of adaptive resonance theory (ART) neural network models used to perform the three primary machine learning modalities, namely, unsupervised, supervised and reinforcement learning. It comprises a representative list from classic to modern ART models, thereby painting a general picture of the architectures developed by researchers over the past 30 years. The learning dynamics of these ART models are briefly described, and their distinctive characteristics such as code representation, long-term memory and corresponding geometric interpretation are discussed. Useful engineering properties of ART (speed, configurability, explainability, parallelization and hardware implementation) are examined along with current challenges. Finally, a compilation of online software libraries is provided. It is expected that this overview will be helpful to new and seasoned ART researchers. 7 Code repositories 54 8 Conclusions 55 2 Compute activation function(s): Tj = fT (x, θj, α), ∀j ∈ C. 3 Perform WTA competition: J = arg max j∈Λ (Tj). 4 Compute match function(s):7 Update category J: θ new J = fL(x, θ old J , β). 8 else 9 Deactivate category J: Λ ← Λ − {J}. 10 if Λ = {∅} then 11 Go to step 3. 12 else 13 Set J = |C| + 1. 14 Create new category: C ← C ∪ {J}. 15 Initialize new category: θ new J = fN (x, λ). 16 Set output: y where the vigilance criterion is M J ≥ ρ.Learning. If the winning category satisfies the vigilance test, then resonance occurs, and the radius R J and centroid m J of the winning node are updated as follows:where ρ ∈ (0, 1] is the vigilance parameter.Learning. If the winning category J satisfies M J ≥ ρ, then resonance occurs, and it is updated as follows:J J needs to be updated after the presentation of each sample, then γ k (new) can be estimated incrementally. Particularly, when there is a resonant committed node J, if x k is a meta-information feature, thenotherwise,

show abstract

Uncertain fuzzy self-organization based clustering: interval type-2 fuzzy approach to adaptive resonance theory

Cited by 13 publications

References 31 publications

An improved de-interleaving algorithm of radar pulses based on SOFM with self-adaptive network topology

An improved de-interleaving algorithm of radar pulses based on SOFM with self-adaptive network topology

Data Clustering Method Using Efficient Fuzzifier Values Derivation

A survey of adaptive resonance theory neural network models for engineering applications

Contact Info

Product

Resources

About