Deep rule-based classifier with human-level performance and characteristics
نویسندگان
چکیده
منابع مشابه
Learning Fuzzy Rule Based Classifier in High Performance Computing Environment
An approach to estimate the number of rules by spectral analysis of the training dataset has been recently proposed [1]. This work presents an analysis of such a method in high performance computing environment. Two approaches for parallel implementation of the method were studied considering the structure selection genetic algorithm and the spectral decomposition. The results show that both ap...
متن کاملFault Detection of Bearings Using a Rule-based Classifier Ensemble and Genetic Algorithm
This paper proposes a reduct construction method based on discernibility matrix simplification. The method works with genetic algorithm. To identify potential problems and prevent complete failure of bearings, a new method based on rule-based classifier ensemble is presented. Genetic algorithm is used for feature reduction. The generated rules of the reducts are used to build the candidate base...
متن کاملeffect of bataine and sulphate supplement on wool and milk characteristics and lambs performance in naine ewes
تعداد 20 رأس میش نژاد نائینی 6+-24 ماهه، با میانگین وزن 2/3-+40 کیلوگرم، همراه با 20 رأس بره های آنها در قالب طرح کاملا تصادفی اثر بتائین و مکمل سولفات بر خصوصیات شیر و پشم و عملکرد بره ها بررسی شد. میش ها بطور تصادفی در چهار تیمار قرار گرفتند. تیمارها عبارت بودند از: 1-گروه شاهد 2-بتائین (05/0 درصد ماده خشک) 3-سولفات (24/0 سولفور درصد ماده خشک 4-بتائین هرماه با سولفات. جیره غذائی طبق nrc با مح...
15 صفحه اولVoltage Sag Compensation with DVR in Power Distribution System Based on Improved Cuckoo Search Tree-Fuzzy Rule Based Classifier Algorithm
A new technique presents to improve the performance of dynamic voltage restorer (DVR) for voltage sag mitigation. This control scheme is based on cuckoo search algorithm with tree fuzzy rule based classifier (CSA-TFRC). CSA is used for optimizing the output of TFRC so the classification output of the network is enhanced. While, the combination of cuckoo search algorithm, fuzzy and decision tree...
متن کاملData characteristics that determine classifier performance
We study the relationship between the distribution of data, on the one hand, and classifier performance, on the other, for non-parametric classifiers. It is shown that predictable factors such as the available amount of training data (relative to the dimensionality of the feature space), the spatial variability of the effective average distance between data samples, and the type and amount of n...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Information Sciences
سال: 2018
ISSN: 0020-0255
DOI: 10.1016/j.ins.2018.06.048