GOFAM: a hybrid neural network classifier combining fuzzy ARTMAP and genetic algorithm

Yaghini, Masoud; Shadmani, M

doi:10.1007/s10462-011-9265-3

Cited by 7 publications

(5 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The incremental learning nature of the FA networks makes them susceptible to ordering effects. Several approaches reported either averaging classifiers with multiple arbitrary training sequences [62] or used an optimization technique such as GA to find the best sequence [40]. In addition, hyperparameter tuning is necessary to ensure that the network can cluster the training data with minimal information loss.…”

Section: Hyper-parameter Optimization Using Genetic Algorithmsmentioning

confidence: 99%

Emotion Recognition Using Explainable Genetically Optimized Fuzzy ART Ensembles

2021

View full text Add to dashboard Cite

There is a growing demand for explainability in complex artificial intelligence solutions to support critical applications' decision-making processes. Barriers to explainable processes include blackbox classifiers, such as deep learning, and noisy datasets. Affect recognition involving neural networks attempts to map complex human emotions onto Arousal and Valence scales based on physiological signal measurements. Datasets collected for this purpose are inherently noisy and may contain outliers and imbalanced classes, hindering accurate classification. In our approach, these issues are addressed using Fuzzy ART (FA) for clustering data samples into more condensed memory templates, introducing stochastic resonant noise to amplify signal-to-noise ratio, and SMOTE sampling to generate synthetic minority samples. A genetic algorithm is developed for FA optimization and ensemble model selection. Clusters obtained from the resulting ensembles are then used to train an ensemble of boosted decision trees for classification and to visualize the decision-making processes. Individual features such as heart rate variability and EEG band power, as well as feature interactions between pairs of features, may contain critical information as human affect indicators. Contributions of individual features and feature interactions toward describing human affect are quantified and interpreted using Shapley additive explanation values. Three established affect recognition datasets were considered for mapping physiological features onto a binary classification of Low/High Arousal and Positive/Negative Valence. Our framework was able to achieve good generalization for both classification tasks as well as provide detailed insights into the contributions of physiological features towards describing Arousal and Valence affects.

show abstract

Section: Hyper-parameter Optimization Using Genetic Algorithmsmentioning

confidence: 99%

Emotion Recognition Using Explainable Genetically Optimized Fuzzy ART Ensembles

2021

View full text Add to dashboard Cite

show abstract

“…The results showed improvements in the emotion recognition rate based on 25 selected features. The GA was introduced in the work of Yaghini and Shadmani to find a good data presentation order for FAM training. The proposed method, ie, genetic‐ordered FAM, was evaluated using several data sets from the University of California, Irvine (UCI) machine learning repository.…”

Section: Literature Reviewmentioning

confidence: 99%

“…To mitigate this issue, the centroid of each FAM prototype node is tagged with a weight, ie, its confidence factor calculated using Equation (5). Based on this weight information, Equation (9) is used to replace Equation (7). Thus, we have…”

Section: Modifications Of Cartmentioning

confidence: 99%

“…One of the models that offers incremental learning is the family of adaptive‐resonance‐theory (ART) neural networks . Among different ART models, the fuzzy ARTMAP (FAM) network, which is one of the most successful neural network classifiers, possesses the incremental learning property and the capability of overcoming the stability‐plasticity dilemma . Over the years, there are a number of extensions to FAM and other fuzzy‐based models, which include fuzzy lattice reasoning techniques and application of fuzzy models to human facial expression recognition, have been proposed.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A hybrid FAM‐CART model for online data classification

Seera

Lim

Tan

2018

Computational Intelligence

View full text Add to dashboard Cite

In this paper, an online soft computing model based on an integration between the fuzzy ARTMAP (FAM) neural network and the classification and regression tree (CART) for undertaking data classification problems is presented. Online FAM network is useful for conducting incremental learning with data samples, whereas the CART model prevails in depicting the knowledge learned explicitly in a tree structure. Capitalizing on their respective advantages, the hybrid FAM‐CART model is capable of learning incrementally while explaining its predictions with knowledge elicited from data samples. To evaluate the usefulness of FAM‐CART, 2 sets of benchmark experiments with a total of 12 problems are used in both offline and online learning modes. The results are examined and compared with those published in the literature. The experimental outcome positively indicates that the online FAM‐CART model is useful for tackling data classification tasks. In addition, a decision tree is produced to allow users in understanding the predictions, which is an important property of the hybrid FAM‐CART model in supporting decision‐making tasks.

show abstract

“…At the present time the nerve cell chip based on ART1 has been produced [6]. Up to now many new ART methods are studied for different application fields [7].…”

Section: Introductionmentioning

confidence: 99%

New ART2/ART2A Algorithm Apply to Entire Real Number Field

Shen

et al. 2013

AMR

View full text Add to dashboard Cite

In this paper, two shortcomings of standard ART2/ART2A algorithm were revealed through theoretical analysis: (1)Standard ART2/ART2A algorithm is only suitable for the case in the nonnegative real number field because of a limit of pretreating process in F1layer; (2)Even through all input patterns are shifted to the nonnegative real number field through coordinate transformation, the standard ART2/ART2A algorithm can not correctly recognize those patterns which have same phase, but different amplitudes. As a result, the standard ART2/ART2A algorithm is not quite suitable for universal pattern recognition. So this paper presented a new nonlinear transforming function in F1layer and a new competitive learning formula in F2layer for traditional ART2/ART2A algorithm. The applicable scope of the new ART2/ART2A algorithm is expanded to entire real number field from nonnegative real number field. The result of typical calculation example shows that the presented algorithm is effective.

show abstract

GOFAM: a hybrid neural network classifier combining fuzzy ARTMAP and genetic algorithm

Cited by 7 publications

References 20 publications

Emotion Recognition Using Explainable Genetically Optimized Fuzzy ART Ensembles

Emotion Recognition Using Explainable Genetically Optimized Fuzzy ART Ensembles

A hybrid FAM‐CART model for online data classification

New ART2/ART2A Algorithm Apply to Entire Real Number Field

Contact Info

Product

Resources

About