Abstract | ||
---|---|---|
The purpose of a classification algorithm is to predict the class label of a new instance based on the analysis of a training dataset. Many classification algorithms work most naturally with nominal attributes. However, numerical data are very common in real-life applications. In this paper, we present a classification learning for numerical datasets. We adopt the idea of function decomposition, which is an approach used to represent a complex function by simple and smaller subfunctions. We modify the decomposition and apply it to decompose numerical datasets for classification learning. The proposed method is also implemented to evaluate the classification accuracy. The experimental evaluation shows the proposed method is a relatively effective method for classification learning. |
Year | DOI | Venue |
---|---|---|
2002 | 10.1007/3-540-45675-9_68 | IDEAL |
Keywords | Field | DocType |
numerical datasets,numerical data,effective method,complex function,classification accuracy,classification algorithm,class label,function decomposition,classification learning,functional decomposition | One-class classification,Effective method,Computer science,Functional decomposition,Decomposition method (constraint satisfaction),Artificial intelligence,Statistical classification,Time complexity,Linear classifier,Machine learning,Multiclass classification | Conference |
Volume | ISSN | ISBN |
2412 | 0302-9743 | 3-540-44025-9 |
Citations | PageRank | References |
0 | 0.34 | 6 |
Authors | ||
2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Grace J. Hwang | 1 | 4 | 1.09 |
Chun-Chan Tung | 2 | 0 | 0.34 |