Analysis of data mining classification by comparison of C4.5 and ID algorithms

The Indonesian Operations Research Association (IORA) - International Conference on Operations Research 2016

Sudrajat, R.^1 ; Irianingsih, I.^1 ; Krisnawan, D.^2

Department of Computer Science, Padjadjaran University, Sumedang

45363, Indonesia^1

Department of Mathematics, Padjadjaran University, Sumedang

45363, Indonesia^2

关键词: Analysis of data;    C4.5 algorithm;    Discrete data;    Error rate;    ID3 algorithm;    Numerical data;    Similar datum;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/166/1/012031/pdf
DOI  :  10.1088/1757-899X/166/1/012031

来源: IOP

【 摘 要 】

The rapid development of information technology, triggered by the intensive use of information technology. For example, data mining widely used in investment. Many techniques that can be used assisting in investment, the method that used for classification is decision tree. Decision tree has a variety of algorithms, such as C4.5 and ID3. Both algorithms can generate different models for similar data sets and different accuracy. C4.5 and ID3 algorithms with discrete data provide accuracy are 87.16% and 99.83% and C4.5 algorithm with numerical data is 89.69%. C4.5 and ID3 algorithms with discrete data provides 520 and 598 customers and C4.5 algorithm with numerical data is 546 customers. From the analysis of the both algorithm it can classified quite well because error rate less than 15%.