Evaluation of classification algorithms for biological data analysis
Due to the rapid growth in the amount of biological data, data mining techniques have become common and efficient tools in extracting and analyzing valuable information within those data. Biological data has several characteristics such as high dimensional, class imbalanced, noisy, making it difficult to be solved by conventional statistical methods. Classification, which is the next step after feature subset selection (Damanik 2008), is a part of the data mining process to allocate objects to one of several categories that have been defined before. Various classification algorithms have been proposed, including nearest neighbor, artificial neural network, and kernel based algorithms. The aim of the present study is to make a comprehensive evaluation of those algorithms, compare their performances and characteristics for data with different scale in term of dimensionality, and to produce important recommendations for application of classification algorithms in biological data mining.
B00641 | (wh) | Available |
No other version available