Download now Free registration required
In this paper, it is tried to implement classifying methods on some given data files and compare the results together in different cases with different parameters. First of all, the authors have a pre-processing step, data normalizing and then the chosen methods are K-Nearest Neighbor (KNN), Probabilistic Neural Network (PNN), K-Nearest Centroid Neighbor (KNCN), Discriminant Adaptive Nearest Neighbor classification (DANN) and Nearest Feature Line (NFL) known as methods for classifying data. KNN definition: In pattern recognition, the k-Nearest Neighbor algorithm (kNN) is a method for classifying objects based on closest training samples in the feature space.
- Format: PDF
- Size: 950.36 KB