International Journal of Computer Applications (0975 – 8887) Volume 11– No.9, December 2010 30 Performances of Estimating Null Values using Noble Evolutionary Algorithm (NEAs) by Generating Weighted Fuzzy Rules Muhammad Firoz Mridha Department of Computer Science Stamford University Bangladesh Dhaka, Bangladesh Manoj Banik Department of Computer Science and Engineering Ahsanullah University of Science and Technology, Dhaka, Bangladesh ABSTRACT This paper Present a noble technique to estimate null values from relational database systems. At present some methods exist to estimate null values from relational database systems. The estimated accuracy of the existing methods are not good enough. We have used an advance technique for estimating null values in relational database systems. In our paper we present the technique to generate weighted Fuzzy rules from relational database systems for estimating null values using Noble Evolutionary algorithms. The parameters (operators) of the Evolutionary algorithms are adapted via Fuzzy systems. We have fuzzified the attribute values using membership functions shape. The results of the evolutionary algorithms are the weights of the attributes. The different weights of attribute generate a set of Fuzzy rules. From this we have obtained a set of rules. Our proposed techniques have a higher average estimated accuracy rate and able to estimate the null values in relational database systems. Keywords Fuzzy System, Membership Functions, Noble Evolutionary algorithms, Null values, Relational Database Systems, Weighted Fuzzy Rules. 1. INTRODUCTION Fuzzy systems have become popular components of consumer products because they are able to solve difficult nonlinear control problem, exhibit robust behavior and present linguistic representations. These rule-based systems are more suitable for complex system problems where it is very difficult to describe the system mathematically. One of the most important considerations in designing any fuzzy system is the generation of the fuzzy rules as well as membership functions for each fuzzy set. This paper present NEAs approach to solve problem. The solving procedure mainly based on Evolutionary algorithms. It has been observed that there are many drawbacks in the early methods in estimating null values. We have estimated null values more accurately as well as to overcome the drawbacks of the previous methods. Global optimization problems are very difficult to solve. In order to understand the difficulties it is important to note that all local optimization techniques can at most locate a local minimum. 2. FUZZY EXPERT SYSTEM The fuzzy expert system works as follows [1]: 1).Determine the fuzzy membership values activated by the inputs.2). Determine which rules are fired in the rule set. 3). Combine the membership values for each activated rule using the AND operator.4). Trace rule activation membership values back through the appropriate output fuzzy membership functions.5). Utilize defuzzification to determine the value for each output variable. 6). Make decision according to the output values. 2.1 Membership Functions A membership function is a curve that defines how each point in the input space is mapped to a membership value as shown in Fig.1 and Fig 2. Fig 1: Membership function for salary attributes Fig.2 Membership function for Experience attributes In this paper, the membership functions of the linguistic term, “L” ,”SL ”,”M” ,”SH” ,and ”H” of the attributes and “Experience” in the relational database system are adopted as shown in Fig. 2[3]. 2.2 Fuzzy Rule Base The general form of a fuzzy rule in a fuzzy system is [1] I f x 11 is S1 , and x 2 is S2 , ... ... ..., x k is Sk Then y 11 is T1 , ... ..., and y1 is T1 2.3. Weighted Fuzzy Rules Weighted fuzzy rules are a set of rules including the weights of the attributes; w ij denotes the weight of attributes A i of the i th rule in the rule