Clustering Analysis of Greedy Heuristic Method in Zero_One Knapsack Problem

Abstract

Knapsack problem is a surely understood class of optimization problems, which tries to expand the profit of items in a knapsack without surpassing its capacity, Knapsack can be solved by several algorithms such like Greedy, dynamic programming, Branch & bound etc. The solution to the zero_one knapsack problem (KP) can be viewed as the result of a sequence of decision. Clustering is the process of resolving that type of applications. Different clustering application for grouping elements with equal priority. In this paper we are introducing greedy heuristic algorithm for solving zero_one knapsack problem. We will exhibit a relative investigation of the Greedy, dynamic programming, B&B and Genetic algorithms regarding of the complexity of time requirements, and the required programming efforts and compare the total value for each of them. Greedy and Genetic algorithms can be used to solve the 0-1 Knapsack problem within a reasonable time complexity. The worst-case time complexity (Big-O) of both algorithms is O(N). Using the greedy method, the algorithm can produce high quality clusters while reduce time the best partioning avoid the memory confinement problem during the process.