Design and Implementation of Collage Plane Model Using Decision Tree and Clustering Algorithms

Abstract

The amount of data kept in computer files and databases are growing at a phenomenal rate .the users of these data are expecting more sophisticated information from them .Simples SQL structure query language queries is not adequate to support these needs. A data-mining algorithm is the mathematical and statistical algorithm that transforms the cases in the original data source into the data-mining model. However the model looks depends largely on the data mining algorithm applied to the. Where a decision tree algorithm is used to analysis the data and creates a repeating series of branches until no more relevant branches be made. The end result is a binary tree structure where the splits in the branches can be followed along specific criteria to find the most desired result .unlike a decision tree, a cluster a algorithm, does not split data along any lines but rather group’s data in clusters. Clustering is most useful for visual representations because the data is ground around common criteria. The general objective is to build trees are as balanced as possible, in terms of the distribution of attributes. in other words the algorithm seeks to put as many of one type of attribute as possible in a given node another commonly used algorithm to create models is clustering . The algorithm creates a model that cannot be used to make predictions but is very effective in finding records that have attributes in common with each other. In this research the decision support for database and data mining algorithms (decision tree and clustering) are explain, the goal of data miming algorithm in decision support is to create rules (for every node which can expressed as a set of rules that provide a description of the function of that particular node in the tree as well as the node that led up to it). We build a system called (DM college plan) that using decision tree and clustering algorithms, the goal is to create module that can be used to finds records that have attributes in common with each other.