Implementation of Data Mining in Estimating The Growth Of Local Sheep

Data mining is a process to use statistical technique, mathematics, artificial intelligence, and learning machine to extract, identify beneficial information and discovery knowledge from database. In this research, the authors apply this method to estimate the growth of local sheep. Research method consists of several phases, namely: Data Cleaning, Data Integration, Data Selection, Data Transformation, Data Mining, Pattern Evolution and Knowledge Presentation. Data as amount of 4357 samples, processed by using CART (Classification and Regression Tree) and Correlation Analysis method. The Average Daily Gain is target variable is and indicator variable consist of dry matter intake from : Grass; Corn; Cassava Meal; Coconut Meal; CaCO3; Salt; Premix; Urea; Corn Oil; Corncob; Soybean Meal; Fish Meal and Sunflower Oil. The knowledge presentation gotten is Coconut Meal as dominant indicator variable. The optimal regression trees that has 41 terminal nodes with relative error of 0,659, can be used to determine composition ingredient base on daily gain expected.