Prototype-Based Threshold Rules
نویسندگان
چکیده
Understanding data is usually done extracting fuzzy or crisp logical rules using neurofuzzy systems, decision trees and other approaches. Prototypebased rules are an interesting alternative providing in many cases simpler, more accurate and more comprehensible description of the data. Algorithm for generation of threshold prototype-based rules are described and a comparison with neurofuzzy systems on a number of datasets provided. Results show that systems for data understanding generating prototypes deserve at least the same attention as that enjoyed by the neurofuzzy systems.
منابع مشابه
Identification and Robust Fault Detection of Industrial Gas Turbine Prototype Using LLNF Model
In this study, detection and identification of common faults in industrial gas turbines is investigated. We propose a model-based robust fault detection(FD) method based on multiple models. For residual generation a bank of Local Linear Neuro-Fuzzy (LLNF) models is used. Moreover, in fault detection step, a passive approach based on adaptive threshold is employed. To achieve this purpose, the a...
متن کاملExtraction of Prototype-Based Threshold Rules Using Neural Training Procedure
Complex neural and machine learning algorithms usually lack comprehensibility. Combination of sequential covering with prototypes based on threshold neurons leads to a prototype-threshold based rule system. This kind of knowledge representation may be quite powerful, providing solutions to many classification problems using a single rule.
متن کاملPrototype Rules from SVM
1 Why prototype-based rules? Propositional logical rules may not be the best way to understand the class structure of data describing some objects or states of nature. The best explanation may differ depending on the problem, the type of questions and the type of explanations that are commonly accepted in a given field. Although most research has focused on propositional logical rules [14, 19] ...
متن کاملThreshold rules decision list
Understanding data is one of most important problems. Popular crisp logic rules are easy to understand and compare, however for some datasets the number of extracted rules is very large, what affect reduction of generalization and makes the system less transparent. Another solution are fuzzy logic rules, which are much more flexible, however they don’t support symbolic and nominal attributes. A...
متن کاملSelection of Prototype Rules: Context Searching Via Clustering
Prototype-based rules are an interesting alternative to fuzzy and crisp logical rules, in many cases providing simpler, more accurate and more comprehensible description of the data. Such rules may be directly converted to fuzzy rules. A new algorithm for generation of prototype-based rules is introduced and a comparison with results obtained by neurofuzzy systems on a number of datasets provided.
متن کامل