Analysis of Iterative Dichotomiser 3 Algorithm Uses Fuzzy Curves Shoulder as a Determinant of Grade Value Arina Prima Silalahi 1 , Zakarias Situmorang 2 , Syahril Efendi 3 , and Eva Darnila 4 {primaarinasilalahi@gmail.com} 1 Magister of Information Technology, Universitas Sumatera Utara, Medan, Indonesia 2 Department of Computer Science, University of Katolik Santo Thomas, Medan, Indonesia 3 Department of Information Technology, Universitas Sumatera Utara, Medan, Indonesia 4 Department of Informatics, Universitas Malikussaleh, Aceh Utara, Indonesia Abstract. Data mining is a process that combines statistics, artificial intelligence, mathematics and machine learning to extract data on a large scale in the database. Data mining is always able to analyze the data so as to find the relevance of data that has a meaning and have a tendency to check large-scale data stored in the database to find a meaningful pattern or rules. The increasing availability of data is often not utilized to provide new knowledge so that large data accumulate is meaningless. The purpose of this research is to extract the information so as to produce knowledge through the decision tree and show the accuracy or influence of Iterative Algorithm Dichotomiser 3 which is used to predict a situation. The classes or attributes in the Iterative Algorithm Dichotomiser are continuously broken into relative categories. Fuzzy Curve Shoulder will be used as a function to form the categories of each attribute value. Using a fuzzy shoulder curve, the dataset is processed using a decision tree that is useful for extracting large amounts of data and searching for hidden links between multiple potential input variables with a target variable. The results of this study are decision trees that will provide predictive data with Iterative Dichotomizer (ID) Algorithm 3. Keywords: Data mining, Fuzzy Curve Shoulder, Iterative Dichotomizer Algorithm. 1 Introduction Knowledge Discovery in Database (KDD) is a method to gain knowledge from existing databases. In the database there are tables-tables related to each other. The knowledge obtained in the process can be used as a knowledge base for decision making purposes. Data mining concerns theories, methodologies, and in particular, computer systems for knowledge extraction or mining from large amounts of data. Data mining is a method to extract the knowledge and information from a large number data such as incomplete, noisy and random [1]. Data mining explorers data from within the database to find hidden patterns, searching for information to predict data. Data Mining techniques are used to examine large databases as a way to discover new and useful patterns. The decision tree is considered one of the most popular approaches. In the decision tree classification consists of a node that forms the root [2]. Decision tree can describe the WMA-1 2018, January 20-21, Indonesia Copyright © 2019 EAI DOI 10.4108/eai.20-1-2018.2281862