A Fuzzy Rule Extraction Method for ANFIS Using CFCM and Fuzzy Equalization

2000 ◽  
Vol 4 (5) ◽  
pp. 355-361 ◽  
Author(s):  
Myung-Geun Chun ◽  
◽  
Keun-Chang Kwak ◽  
Jeong-Woong Ryu ◽  
Witold Pedrycz ◽  
...  
Keyword(s):  
Fuzzy Inference ◽  
Fuzzy Rule ◽  
Fuzzy Data ◽  
Rule Generation ◽  
Membership Functions ◽  
Fe Method ◽  
Inference System ◽  
Fuzzy Membership Functions ◽  
Satisfactory Performance ◽  
The Given

In this paper, an efficient fuzzy rule generation scheme for Adaptive Network-based Fuzzy Inference System (ANFIS) using the conditional fuzzy c-means (CFCM) and fuzzy equalization (FE) methods is proposed. Here, the CFCM is adopted to render clusters, which can represent the homogeneous properties of the given input and output fuzzy data. And also the FE method is used to automatically construct the fuzzy membership functions for ANFIS. From this, we can systematically obtain a small size of fuzzy rules that shows satisfactory performance for the given problems. We applied the proposed method to the truck-backing control and Box-Jenkins modeling problems and obtained a better result than previous work.

A Fuzzy Inference System for Identifying Tissue Elasticity Using Ultrasound

2003 ◽  
Vol 7 (1) ◽  
pp. 31-39 ◽  
Author(s):  
Tadashi Kimura ◽  
◽  
Kouki Nagamune ◽  
Syoji Kobashi ◽  
Katsuya Kondo ◽  
...  
Keyword(s):  
Fuzzy Inference ◽  
Fuzzy Rule ◽  
Measurement Range ◽  
Identification System ◽  
Membership Functions ◽  
Tissue Elasticity ◽  
Inference System ◽  
Fuzzy Membership Functions ◽  
Rule Based Approach ◽  
Two Stages

This paper proposes a fuzzy rule-based approach for identifying tissue elasticity using ultrasound. The purpose of this paper identifies automatically tissue elasticity. Information of tissue elasticity helps us to diagnose several diseases. Elastography was able to estimate tissue elasticity. However, this measurement range is limited due to the need of pressure. To avoid this limitation, this paper proposes the identification system without pressure. This inference system consists of two stages. In the first stage, fuzzy membership functions are constructed by known data of elasticity. The second stage identifies elasticity of unknown data by using the membership functions. We used five different phantoms (total 5×10 = 50) of elasticity as known data and applied this system into nine different phantoms (total 9×10 = 90) of elasticity as unknown data. As a result, the correlation coefficient between actual value and identified value was 0.789 and the error of means was 0.646. This system thus acquired smaller error ratio than that of the statistical method.

scholarly journals Tuning of a TS Fuzzy Output Regulator Using the Steepest Descent Approach and ANFIS

2013 ◽  
Vol 2013 ◽  
pp. 1-14 ◽  
Author(s):  
Ricardo Tapia-Herrera ◽  
Jesús Alberto Meda-Campaña ◽  
Samuel Alcántara-Montes ◽  
Tonatiuh Hernández-Cortés ◽  
Lizbeth Salgado-Conrado
Keyword(s):  
Fuzzy Inference ◽  
Fuzzy Controller ◽  
Descent Method ◽  
Output Regulation ◽  
Steepest Descent ◽  
Steepest Descent Method ◽  
Membership Functions ◽  
Inference System ◽  
Fuzzy Membership Functions ◽  
Takagi Sugeno

The exact output regulation problem for Takagi-Sugeno (TS) fuzzy models, designed from linear local subsystems, may have a solution if input matrices are the same for every local linear subsystem. Unfortunately, such a condition is difficult to accomplish in general. Therefore, in this work, an adaptive network-based fuzzy inference system (ANFIS) is integrated into the fuzzy controller in order to obtain the optimal fuzzy membership functions yielding adequate combination of the local regulators such that the output regulation error in steady-state is reduced, avoiding in this way the aforementioned condition. In comparison with the steepest descent method employed for tuning fuzzy controllers, ANFIS approximates the mappings between local regulators with membership functions which are not necessary known functions as Gaussian bell (gbell), sigmoidal, and triangular membership functions. Due to the structure of the fuzzy controller, Levenberg-Marquardt method is employed during the training of ANFIS.

Comparison of Mamdani and Sugeno Inference Methods in Calorie Burn Calculation for Activity Using Treadmill

2020 ◽  
Vol 17 (4) ◽  
pp. 1703-1709
Author(s):  
Prabhakar Rontala Subramaniam ◽  
Chitra Venugopal
Keyword(s):  
Linear System ◽  
Membership Function ◽  
Mathematical Analysis ◽  
Fuzzy Inference ◽  
Fuzzy Rule ◽  
Membership Functions ◽  
Inference System ◽  
Calculation Results ◽  
Inference Methods ◽  
Better Than

This research provides a comparison between the performance of Mamdani and Sugeno Fuzzy Inference System in Calorie Burn calculation for treadmill workout exercise. The results are compared against the calculation calorie burn value using traditional formula. The objective of this results to get a system which provides calorie burn results closer to the traditionally calculated value. The Mamdani and Sugeno systems are designed with two input membership functions namely Incline and Speed and one output membership function called Calorie Burn as used in the traditional calculation method. The fuzzy rule table is designed using 7 Membership function for Incline and speed inputs and 9 Membership function for calorie burn output. The Mamdani and Sugeno system are tested with the same fuzzy rules for better comparison. The results are tabulated with theoretical values compared with Mamdani and Sugeno system for 5 incline levels and 7 speed levels. It can be observed that the Sugeno system results are closer to the traditional calculation results. This is shows that the sugeno system works better than Mamdani system for linear system. Also it depicts the calculated values and hence it can be used as a replacement for tedious mathematical analysis.

scholarly journals Approximate reasoning with fuzzy rule interpolation: background and recent advances

2021 ◽  
Author(s):  
Fangyi Li ◽  
Changjing Shang ◽  
Ying Li ◽  
Jing Yang ◽  
Qiang Shen
Keyword(s):  
Fuzzy Inference ◽  
Fuzzy Rule ◽  
Approximate Reasoning ◽  
Rule Based ◽  
Reasoning Systems ◽  
Significant Interest ◽  
Recent Developments ◽  
Rule Bases ◽  
The Given ◽  
Computing Approach

AbstractApproximate reasoning systems facilitate fuzzy inference through activating fuzzy if–then rules in which attribute values are imprecisely described. Fuzzy rule interpolation (FRI) supports such reasoning with sparse rule bases where certain observations may not match any existing fuzzy rules, through manipulation of rules that bear similarity with an unmatched observation. This differs from classical rule-based inference that requires direct pattern matching between observations and the given rules. FRI techniques have been continuously investigated for decades, resulting in various types of approach. Traditionally, it is typically assumed that all antecedent attributes in the rules are of equal significance in deriving the consequents. Recent studies have shown significant interest in developing enhanced FRI mechanisms where the rule antecedent attributes are associated with relative weights, signifying their different importance levels in influencing the generation of the conclusion, thereby improving the interpolation performance. This survey presents a systematic review of both traditional and recently developed FRI methodologies, categorised accordingly into two major groups: FRI with non-weighted rules and FRI with weighted rules. It introduces, and analyses, a range of commonly used representatives chosen from each of the two categories, offering a comprehensive tutorial for this important soft computing approach to rule-based inference. A comparative analysis of different FRI techniques is provided both within each category and between the two, highlighting the main strengths and limitations while applying such FRI mechanisms to different problems. Furthermore, commonly adopted criteria for FRI algorithm evaluation are outlined, and recent developments on weighted FRI methods are presented in a unified pseudo-code form, easing their understanding and facilitating their comparisons.

ANFIS Model for Time Series Prediction

2013 ◽  
Vol 385-386 ◽  
pp. 1411-1414 ◽  
Author(s):  
Xue Bo Jin ◽  
Jiang Feng Wang ◽  
Hui Yan Zhang ◽  
Li Hong Cao
Keyword(s):  
Time Series ◽  
Fuzzy Inference System ◽  
Prediction Error ◽  
Fuzzy Inference ◽  
Time Series Prediction ◽  
Prediction Performance ◽  
Membership Functions ◽  
Adaptive Network ◽  
Anfis Model ◽  
Inference System

This paper describes an architecture of ANFIS (adaptive network based fuzzy inference system), to the prediction of chaotic time series, where the goal is to minimize the prediction error. We consider the stock data as the time series. This paper focuses on how the stock data affect the prediction performance. In the experiments we changed the number of data as input of the ANFIS model, the type of membership functions and the desired goal error, thereby increasing the complexity of the training.

Finding Active Membership Functions for Genetic-Fuzzy Data Mining

2015 ◽  
Vol 14 (06) ◽  
pp. 1215-1242 ◽  
Author(s):  
Chun-Hao Chen ◽  
Tzung-Pei Hong ◽  
Yeong-Chyi Lee ◽  
Vincent S. Tseng
Keyword(s):  
Data Mining ◽  
Association Rules ◽  
Fuzzy Data ◽  
Membership Functions ◽  
Linguistic Terms ◽  
Fuzzy Association Rules ◽  
Fitness Functions ◽  
The Given ◽  
Binary Strings ◽  
Active Membership

Since transactions may contain quantitative values, many approaches have been proposed to derive membership functions for mining fuzzy association rules using genetic algorithms (GAs), a process known as genetic-fuzzy data mining. However, existing approaches assume that the number of linguistic terms is predefined. Thus, this study proposes a genetic-fuzzy mining approach for extracting an appropriate number of linguistic terms and their membership functions used in fuzzy data mining for the given items. The proposed algorithm adjusts membership functions using GAs and then uses them to fuzzify the quantitative transactions. Each individual in the population represents a possible set of membership functions for the items and is divided into two parts, control genes (CGs) and parametric genes (PGs). CGs are encoded into binary strings and used to determine whether membership functions are active. Each set of membership functions for an item is encoded as PGs with real-number schema. In addition, seven fitness functions are proposed, each of which is used to evaluate the goodness of the obtained membership functions and used as the evolutionary criteria in GA. After the GA process terminates, a better set of association rules with a suitable set of membership functions is obtained. Experiments are made to show the effectiveness of the proposed approach.

scholarly journals Underground Risk Index Assessment and Prediction Using a Simplified Hierarchical Fuzzy Logic Model and Kalman Filter

Processes ◽  
2018 ◽  
Vol 6 (8) ◽  
pp. 103 ◽  
Author(s):  
Muhammad Fayaz ◽  
Israr Ullah ◽  
Do-Hyeun Kim
Keyword(s):  
Fuzzy Logic ◽  
Kalman Filter ◽  
Fuzzy Inference ◽  
Risk Index ◽  
Significant Risk ◽  
Assessment Method ◽  
Logic Model ◽  
Membership Functions ◽  
Inference System ◽  
Fuzzy Logic Model

Normally, most of the accidents that occur in underground facilities are not instantaneous; rather, hazards build up gradually behind the scenes and are invisible due to the inherent structure of these facilities. An efficient inference system is highly desirable to monitor these facilities to avoid such accidents beforehand. A fuzzy inference system is a significant risk assessment method, but there are three critical challenges associated with fuzzy inference-based systems, i.e., rules determination, membership functions (MFs) distribution determination, and rules reduction to deal with the problem of dimensionality. In this paper, a simplified hierarchical fuzzy logic (SHFL) model has been suggested to assess underground risk while addressing the associated challenges. For rule determination, two new rule-designing and determination methods are introduced, namely average rules-based (ARB) and max rules-based (MRB). To determine efficient membership functions (MFs), a module named the heuristic-based membership functions allocation (HBMFA) module has been added to the conventional Mamdani fuzzy logic method. For rule reduction, a hierarchical fuzzy logic model with a distinct configuration has been proposed. In the simplified hierarchical fuzzy logic (SHFL) model, we have also tried to minimize rules as well as the number of levels of the hierarchical structure fuzzy logic model. After risk index assessment, the risk index prediction is carried out using a Kalman filter. The prediction of the risk index is significant because it could help caretakers to take preventive measures in time and prevent underground accidents. The results indicate that the suggested technique is an excellent choice for risk index assessment and prediction.

Enhancing Fuzzy Inference System-Based Criterion-Referenced Assessment with a Similarity Reasoning Technique

2012 ◽  
pp. 302-327
Author(s):  
Tze Ling Jee ◽  
Kai Meng Tay ◽  
Chee Khoon Ng
Keyword(s):  
Fuzzy Inference System ◽  
Analogical Reasoning ◽  
Fuzzy Inference ◽  
Fuzzy Rule ◽  
Rule Base ◽  
Large Set ◽  
Inference System ◽  
Similarity Reasoning ◽  
Criterion Referenced ◽  
Reasoning Technique

A search in the literature reveals that the use of fuzzy inference system (FIS) in criterion-referenced assessment (CRA) is not new. However, literature describing how an FIS-based CRA can be implemented in practice is scarce. Besides, for an FIS-based CRA, a large set of fuzzy rules is required and it is a rigorous work in obtaining a full set of rules. The aim of this chapter is to propose an FIS-based CRA procedure that incorporated with a rule selection and a similarity reasoning technique, i.e., analogical reasoning (AR) technique, as a solution for this problem. AR considers an antecedent with an unknown consequent as an observation, and it deduces a conclusion (as a prediction of the consequent) for the observation based on the incomplete fuzzy rule base. A case study conducted in Universiti Malaysia Sarawak is further reported.

Fuzzy-Rule Based Adaptive Data Warehouse

2012 ◽  
Vol 3 (1) ◽  
pp. 47-65 ◽  
Author(s):  
Rajdev Tiwari ◽  
Anubhav Tiwari ◽  
Manu Pratap Singh
Keyword(s):  
Decision Making ◽  
Fuzzy Inference ◽  
Knowledge Workers ◽  
Fuzzy Rule ◽  
Data Warehouses ◽  
Large Reservoir ◽  
Inference System ◽  
Rule Based ◽  
Neuro Fuzzy ◽  
New Generation

Data Warehouses (DWs) are aimed to empower the knowledge workers with information and knowledge which helps them in decision making. Technically, the DW is a large reservoir of integrated data that does not provide the intelligence or the knowledge demanded by users. The burden of data analysis and extraction of information and knowledge from integrated data still lies upon the analyst’s shoulder. The overhead of analysts can be taken off by architecting a new generation data warehouses systems those shall be capable of capturing, organizing and representing knowledge along with the data and information in it. This new generation DW may be called as Knowledge Warehouse (KW) shall exhibit decision making capabilities themselves and can also supplement the Decision Support Systems (DSS) in making decisions quickly and effortlessly. This paper proposes and simulates a fuzzy-rule based adaptive knowledge warehouse with capabilities to learn and represent implicit knowledge by means of adaptive neuro fuzzy inference system (ANFIS).

Sign in / Sign up

Export Citation Format

Share Document