A Novel Neuro-Fuzzy Inference System with Multi-Level Membership Function for Classification Applications

2007 ◽  
Vol 11 (4) ◽  
pp. 365-372 ◽  
Author(s):  
Cheng-Jian Lin ◽  
◽  
Chi-Yung Lee ◽  
Cheng-Hung Chen ◽  
Keyword(s):  
Membership Function ◽  
Fuzzy Inference System ◽  
Clustering Algorithm ◽  
Fuzzy Inference ◽  
Learning Algorithm ◽  
Layer Structure ◽  
Inference System ◽  
Neuro Fuzzy ◽  
Multi Level ◽  
Takagi Sugeno

In this paper, a novel neuro-fuzzy inference system with multi-level membership function (NFIS_MMF) for classification applications is proposed. The NFIS_MMF model is a five-layer structure, which combines the traditional Takagi-Sugeno-Kang (TSK). Layer 2 of the NFIS_MMF model contains multi-level membership functions, which are multilevel activation functions. A self-constructing learning algorithm, which consists of the self-clustering algorithm (SCA), fuzzy entropy, and the backpropagation algorithm, is also proposed to construct the NFIS_MMF model and perform parameter learning. Simulations were conducted to show the performance and applicability of the proposed model.

scholarly journals The applicability of Generic Self-Evolving Takagi-Sugeno-Kang neuro-fuzzy model in modeling rainfall–runoff and river routing

2019 ◽  
Vol 50 (4) ◽  
pp. 991-1001 ◽  
Author(s):  
Mohammad Ashrafi ◽  
Lloyd H. C. Chua ◽  
Chai Quek
Keyword(s):  
Fuzzy Inference System ◽  
Hydrological Modeling ◽  
Fuzzy Inference ◽  
Training Dataset ◽  
Rule Base ◽  
Rainfall Runoff ◽  
Inference System ◽  
Neuro Fuzzy ◽  
Takagi Sugeno ◽  
River Routing

Abstract Recent advancements in neuro-fuzzy models (NFMs) have made possible the implementation of dynamic rule base systems. This is in comparison with static applications commonly seen in global NFMs such as the Adaptive-Network-Based Fuzzy Inference System (ANFIS) model widely used in hydrological modeling. This study underlines key differences between local and global NFMs with an emphasis on rule base dynamics, in the context of two common flow forecast applications. A global NFM, ANFIS, and two local NFMs, Dynamic Evolving Neural-Fuzzy Inference System (DENFIS) and Generic Self-Evolving Takagi-Sugeno-Kang (GSETSK), were tested. Results from all NFMs compared favorably when benchmarked against physically based models. Rainfall–runoff modeling is a complex process which benefits from the advanced rule generation and pruning mechanisms in GSETSK, resulting in a more compact rule base. Although ANFIS resulted in the same number of rules, this came about at the expense of having the need for a large training dataset. All NFMs generated a similar number of rules for the river routing application, although local NFMs yielded better results for forecasts at longer lead times. This is attributed to the fact that the routing procedure is less complex and can be adequately modeled by static NFMs.

HUMAN ACTION RECOGNITION USING META-COGNITIVE NEURO-FUZZY INFERENCE SYSTEM

2012 ◽  
Vol 22 (06) ◽  
pp. 1250028 ◽  
Author(s):  
K. SUBRAMANIAN ◽  
S. SURESH
Keyword(s):  
Action Recognition ◽  
Fuzzy Inference System ◽  
Fuzzy Inference ◽  
Learning Algorithm ◽  
Human Action Recognition ◽  
Recognition System ◽  
Human Action ◽  
Sequential Learning ◽  
Inference System ◽  
Neuro Fuzzy

We propose a sequential Meta-Cognitive learning algorithm for Neuro-Fuzzy Inference System (McFIS) to efficiently recognize human actions from video sequence. Optical flow information between two consecutive image planes can represent actions hierarchically from local pixel level to global object level, and hence are used to describe the human action in McFIS classifier. McFIS classifier and its sequential learning algorithm is developed based on the principles of self-regulation observed in human meta-cognition. McFIS decides on what-to-learn, when-to-learn and how-to-learn based on the knowledge stored in the classifier and the information contained in the new training samples. The sequential learning algorithm of McFIS is controlled and monitored by the meta-cognitive components which uses class-specific, knowledge based criteria along with self-regulatory thresholds to decide on one of the following strategies: (i) Sample deletion (ii) Sample learning and (iii) Sample reserve. Performance of proposed McFIS based human action recognition system is evaluated using benchmark Weizmann and KTH video sequences. The simulation results are compared with well known SVM classifier and also with state-of-the-art action recognition results reported in the literature. The results clearly indicates McFIS action recognition system achieves better performances with minimal computational effort.

scholarly journals PREDIKSI TINGGI MUKA AIR BENDUNGAN RIAM KANAN MENGGUNAKAN ADAPTIVE NEURO FUZZY INFERENCE SYSTEM

2018 ◽  
Vol 9 (1) ◽  
pp. 11
Author(s):  
Yusri Ikhwani
Keyword(s):  
Membership Function ◽  
Fuzzy Inference System ◽  
Fuzzy Inference ◽  
Inference System ◽  
Neuro Fuzzy

Bendungan riam kanan yang berada kabupaten banjar ialah salah satu waduk terbesar di kalimantan selatan yang ada di aranio, kabupaten banjar. Waduk buatan yang dalam pembangunannya memakan waktu selama 10 tahun ini dibangun membendung 8 sungai yang bersumber dari Pegunungan Meratus. Tujuan utama dibangunnya waduk riam kanan adalah untuk membangun pembangkit listrik tenaga air untuk daerah kalimantan selatan dan sekitarnya.Tujuan penelitian ini ialah untuk memprediksi tinggi muka air bendungan riam kanan menggunakan metode Adaptive Neuro Fuzzy Inference System (ANFIS) agar dapat bermanfaat dalam kebijakan strategis ketahanan energi khususnya ketahanan pangan dan energi listrik, khususnya ketersediaan air untuk saluran irigasi.Perkiraan prediksi ini menggunakan data tinggi muka air bendungan riam kanan dari tahun 2009 sampai dengan 2015 yang didapatkan dari PLTU riam kanan provinsi kalimantan selatan. Prosedur memprediksi diawali dengan melakukan proses pembagian data, yaitu menjadi data pelatihan dan data pengujian. Setelah itu dilakukan penentuan variabel-variabel pendukung input yang memberikan korelasi cukup signifikan terhadap variabel output. Serelah itu melakukan proses pengujian dengan membandingkan 2 membership function untuk menentukan yang mana memiliki tingkat akurasi yang baik dan nilai error yang rendah dalam memprediksi tinggi muka air bendungan riam kanan.Hasilnya ialah prediksi tinggi muka air bendungan riam kanan menggunakan metode Adaptive Neuro Fuzzy Inference System (ANFIS) dengan membandingkan 2 membership function dengan tingkat keakuratan menghasilkan nilai RMSE 0,010065 pada membership function Bell Kata kunci: bendungan riam kanan, anfis, prediksi, tinggi muka air, membership fungtion

scholarly journals Rainfall Forecasting in Banyuwangi Using Adaptive Neuro Fuzzy Inference System

2017 ◽  
Vol 1 (2) ◽  
pp. 65 ◽  
Author(s):  
Gusti Ahmad Fanshuri Alfarisy ◽  
Wayan Firdaus Mahmudy
Keyword(s):  
Membership Function ◽  
Fuzzy Inference System ◽  
Fuzzy Inference ◽  
Hybrid Learning ◽  
Learning Method ◽  
Rainfall Forecasting ◽  
Inference System ◽  
Previous Event ◽  
Neuro Fuzzy ◽  
High Prediction

Rainfall forcasting is a non-linear forecasting process that varies according to area and strongly influenced by climate change. It is a difficult process due to complexity of rainfall trend in the previous event and the popularity of Adaptive Neuro Fuzzy Inference System (ANFIS) with hybrid learning method give high prediction for rainfall as a forecasting model. Thus, in this study we investigate the efficient membership function of ANFIS for predicting rainfall in Banyuwangi, Indonesia. The number of different membership functions that use hybrid learning method is compared. The validation process shows that 3 or 4 membership function gives minimum RMSE results that use temperature, wind speed and relative humidity as parameters.

scholarly journals Identifikasi Gangguan Neurologis Menggunakan Metode Adaptive Neuro Fuzzy Inference System (ANFIS)

2015 ◽  
Vol 9 (2) ◽  
pp. 187
Author(s):  
Jani Kusanti ◽  
Sri Hartati
Keyword(s):  
Ischemic Stroke ◽  
Ct Scan ◽  
Fuzzy Inference System ◽  
Fuzzy Inference ◽  
System Model ◽  
Cluster Center ◽  
Inference System ◽  
Neuro Fuzzy ◽  
Global Threshold ◽  
Takagi Sugeno

AbstrakPenggunaan metode Adaptive Neuro Fuzzy Inference System (ANFIS) dalam proses identifikasi salah satu gangguan neurologis pada bagian kepala yang dikenal dalam istilah kedokteran stroke ischemic dari hasil ct scan kepala dengan tujuan untuk mengidentifikasi lokasi  yang terkena stroke ischemik. Langkah-langkah yang dilakukan dalam proses identifikasi antara lain ekstraksi citra hasil ct scan kepala dengan menggunakan histogram. Citra hasil proses histogram ditingkatkan intensitas hasil citranya dengan menggunakan threshold otsu sehingga didapatkan hasil pixel yang diberi nilai 1 berkaitan dengan obyek sedangkan pixel yang diberi nilai 0 berkaitan dengan background. Hasil pengukuran digunakan untuk proses clustering image, untuk proses cluster image digunakan fuzzy c-mean (FCM). Hasil clustering merupakan deretan pusat cluster, hasil  data digunakan untuk membangun fuzzy inference system (FIS). Sistem inferensi fuzzy yang diterapkan adalah inferensi fuzzy model Takagi-Sugeno-Kang. Dalam penelitian ini ANFIS digunakan untuk mengoptimalkan hasil penentuan lokasi penyumbatan stroke ischemic. Digunakan recursive least square estimator (RLSE) untuk pembelajaran. Hasil RMSE yang didapat pada proses pelatihan sebesar 0.0432053, sedangkan pada proses pengujian dihasilkan tingkat akurasi sebesar 98,66% Kata kunci—stroke ischemik, Global threshold, Fuzzy Inference System model Sugeno, ANFIS, RMSE  Abstract            The use of Adaptive Neuro Fuzzy Inference System (ANFIS) methods in the process of identifying one of neurological disorders in the head, known in medical terms ischemic stroke from the ct scan of the head in order to identify the location of ischemic stroke. The steps are performed in the extraction process of identifying, among others, the image of the ct scan of the head by using a histogram. Enhanced image of the intensity histogram image results using Otsu threshold to obtain results pixels rated 1 related to the object while pixel rated 0 associated with the measurement background. The result used for image clustering process, to process image clusters used fuzzy c-mean (FCM) clustering result is a row of the cluster center, the results of the data used to construct a fuzzy inference system (FIS). Fuzzy inference system applied is fuzzy inference model of Takagi-Sugeno-Kang. In this study ANFIS is used to optimize the results of the determination of the location of the blockage ischemic stroke. Used recursive least squares estimator (RLSE) for learning. RMSE results obtained in the training process of 0.0432053, while in the process of generated test accuracy rate of 98.66% Keywords— Stroke Ischemik, Global threshold, Fuzzy Inference System model Sugeno, ANFIS, RMSE 

Sign in / Sign up

Export Citation Format

Share Document