Investigation of the effectiveness of metric classification methods in identifying attacks in VANET

This paper discusses the problem of improving the efficiency of metric machine learning methods of identification attacks in vehicular adhoc networks (VANETs). The main idea of this research is to select the type of nonlinear functions for calculating the distances between the objects of the sample, describing the traffic of VANET using metric methods, such as the method of k-nearest neighbour with linearly decreasing weights and the Parzen window method. The analysis of the effectiveness of the methods considered was carried out on a synthetically generated sample with three different types of attacks on the network. Computational experiments have shown that the k-nearest neighbour method with decreasing weights based on an exponential function with base a < 1 is more efficient than the Parzen window method by about 0.3% and has an accuracy of 84.15%.

Blended Models for Nearest Neighbour Algorithms for High Dimensional Smart Medical Data

Nearest neighbor algorithms like kNN and Parzen Window are generative algorithms that are used extensively for medical diagnosis and classification of diseases. The data generated or collected in healthcare is high dimensional and cannot be assumed to follow a particular distribution. The conventional approaches fail due to computational complexity, curse of dimensionality, and varying distributions. Hence, this chapter deals with a blending technique for evaluation of nearest neighbor algorithms based on various parameters such as the size of data, dimensions of data, window size, and number of nearest neighbors to make it suitable for massive datasets. Dimensionality reduction and clustering are combined with nearest neighbor classifier such as kNN and Parzen Window to observe the performance of the blended models on various types of datasets. Experimental results on 15 real datasets with various models reveal the efficacy of the proposed blends.