Fault Detection and Classification Technique for HVDC Transmission Lines Using KNN

A Sensor Data Fusion System Based on k-Nearest Neighbor Pattern Classification for Structural Health Monitoring Applications

“…Some of them—fine, medium and coarse k -NN algorithms—make use of the Euclidean distance to determine the nearest neighbors. According to MATLAB, each classifier works as follows [26]: Fine k -NN: A nearest neighbor classifier that makes finely detailed distinctions between classes with the number of neighbors set to one.Medium k -NN: A nearest neighbor classifier with fewer distinctions than a fine k -NN with the number of neighbors set to 10.Coarse k -NN: A nearest neighbor between classes, with the number of neighbors set to 100.Cosine k -NN: A nearest neighbor classifier that uses the cosine distance metric. The cosine distance between two vectors u and v is defined as: $\begin{array}{c}\hfill 1-{\displaystyle \frac{u·v}{\left|u\right|·\left|v\right|}},\end{array}$ that is, one minus the ratio of the inner product of u and v over the product of the norms of u and v .…”

“…One of the most used algorithms in machine learning applications is the k -NN, also known as k -nearest neighbors. k -NN is outstanding due to its simplicity, and the excellent results obtained when this technique is applied to diverse problems [ 26 ]. This algorithm works by using an input vector with the k closest training samples in the feature space.…”

“…To perform the classification, the algorithm identifies the most common class among the k nearest neighbors. The algorithm requires a training to define the neighbors based on the distance from the test sample and a testing step to determine the class to which this test sample belongs [ 26 ].…”

“…Some of them—fine, medium and coarse k -NN algorithms—make use of the Euclidean distance to determine the nearest neighbors. According to MATLAB, each classifier works as follows [ 26 ]: Fine k -NN: A nearest neighbor classifier that makes finely detailed distinctions between classes with the number of neighbors set to one. Medium k -NN: A nearest neighbor classifier with fewer distinctions than a fine k -NN with the number of neighbors set to 10.…”

Automatic Detection of Faults in Race Walking: A Comparative Analysis of Machine-Learning Algorithms Fed with Inertial Sensor Data

“…A kNN algorithm is one of the simplest classification algorithms for activity recognition [34]. Each combination of measured features represents a point in an n-dimensional space, and the classification process is performed by identifying the most common class among the k-nearest neighbors by maximizing the distance among neighbors related to different classes.…”