Search

An interval type-2 fuzzy logic system (IT2FLS) can function well with uncertain data, with which a type-1 fuzzy logic system (T1FLS) is ineffective because its membership function rests upon crisp values. However, similar to T1FLSs, there are challenges associated with IT2FLSs in selecting parameters, which can significantly affect the accuracy of the classification results with their relatively high sensitivity. This paper discusses and proposes a hybrid model based on IT2FLS and particle swarm optimization (PSO) for prediction problems. The main objective of this paper is to find the optimal solution for the unknown fuzzy systems using labelled data for training the fuzzy system. The PSO technique was used to find the optimal parameters of the Gaussian membership functions which utilized for IT2FLSs. The authors tested two data sets for each of the two prediction problems, namely: burnt forest area prediction and wine quality prediction. The predictive results were compared with other predictive methods including random forest (RF), support vector machines (SVM), artificial neural network (ANN), adaptive neuro fuzzy inference system (ANFIS) and IT2FLS with parameters generated by using the fuzzy c-means algorithm (IT2FLS-FCM). Experiment results showed that the proposed method could significantly improve accuracy compared to several other predictive techniques. [ABSTRACT FROM AUTHOR]

Multiplex Fluorescent In Situ Hybridization (M-FISH) is a multi-channel chromosome image generating technique that allows colors of the human chromosomes to be distinguished. In this technique, all chromosomes are labelled with 5 fluors and a fluorescent DNA stain called DAPI (4 in, 6-Diamidino-2-phenylindole) that attaches to DNA and labels all chromosomes. Therefore, a M-FISH image consists of 6 images, and each image is the response of the chromosome to a particular fluor. In this paper, we propose a genetic interval type-2 fuzzy c-means (GIT2FCM) algorithm, which is developed and applied to the segmentation and classification of M-FISH images. Chromosome pixels from the DAPI channel are segmented by GIT2FCM into two clusters, and these chromosome pixels are used as a mask for the remaining five channels. Then, the GIT2FCM algorithm is applied to classify the chromosome pixels into 24 classes, which correspond to the 22 pairs of homologous chromosomes and two sexual chromosomes. The experiments performed using the M-FISH dataset show the advantages of the proposed algorithm. [ABSTRACT FROM AUTHOR]