Influence of Dataset Character on Classification Performance of Support Vector Machines for Grain Analysis

K. Anding, G. Linβ, and P. Brückner (Germany)


machine learning, dataset character, support vector machine, grain analysis


The character of the dataset has a strong influence on the classification performance of each classification task. So it is very important to specify the requirements of a suitable dataset for an optimal classification performance. There is a large number of influencing factors on dataset character. Some of these factors are to be highlighted by the following investigations into the complex application of determination of grain sample components. The recognition of such a natural product is a very challenging and complex task because of its great variability inside each class and a great similarity of some classes. An automated object recognition routine for grain is the task to be solved in an optimal way.

Important Links:

Go Back