What is it about?
In machine learning, fuzzy k-NN classification assigns weights to nearest neighbors samples for interpolating their membership functions to determine the membership function at the position of an unknown (test) sample. We have demonstrated that each of these weights should be based on the geometrical relation among the nearest neighbor, its most informative known neighbor of the same class and the unknown sample.
Photo by Nina Strehl on Unsplash
Why is it important?
Not only is the accuracy of the proposed method higher than the two variants (crisp and fuzzy initialization) of the original fuzzy-kNN method of Keller, but also the proposed method is highly competitive with the best of the fuzzy-kNN type algorithms investigated since then.
Read the Original
This page is a summary of: Fuzzy k-NN classification with weights modified by most informative neighbors of nearest neighbors, Journal of Intelligent & Fuzzy Systems, June 2019, IOS Press, DOI: 10.3233/jifs-18974.
You can read the full text:
The following have contributed to this page