# A Flexible hybrid Genetic-Based Fuzzy Rough Decision Model for accuracy enhancement

### EasyChair Preprint 2045

13 pages•Date: November 28, 2019### Abstract

In this paper, we introduce a new more powerful hybrid algorithm which integrates the advantages of rough set theory and fuzzy set theory together with Genetic Algorithms (GAs). Our Genetic-Based Fuzzy Rough Decision Model algorithm consists of four phases: (1) automatic attributes fuzzification, (2) Eliminate redundant attributes using rough set theory, (3)** **Generating Fuzzy rough rules then calculate automatically the accuracy and a fitness value (Confidence) for each rule and , (4) Using the genetic algorithm for the Fuzzy rough rules to enhance their accuracy. In phase one, the user input the number of fuzzy sets of each attributes, our algorithm determine the maximum and minimum values of each attribute then calculates automatically the width (∆) which divides the universe of discourse of each attribute into “n” intervals according to the number of fuzzy sets, also the algorithm calculates automatically the width (δ_{i}) according to the width (∆). In phase two, we use the rough set techniques to reduce the number of attributes that comes from phase one and produce fuzzy-rough rules. In phase three, the algorithm calculates the accuracy and the confidence (fitness value) of each fuzzy rough rule then calculates the total accuracy of all linguistic rules. In phase four, we run the genetic algorithm on the fuzzy-rough rules from phase three then the algorithm calculates the accuracy and the confidence of each fuzzy rough rule again and calculates the total the accuracy of all rules. The accuracy of our algorithm that applied on Iris plants dataset before using our genetic algorithm from randomly 75 rows from 150 rows is 0.56 but after using our algorithm will be 0.95.

**Keyphrases**: Accuracy, Automated Fuzzy Based Rough Decision model, Fuzzy Logic, Genetic Algorithm, rough set