TITLE:
Applying K-Means Clustering and Fuzzy C-Means Clustering in Vehicle Crashes
AUTHORS:
Azad Abdulhafedh
KEYWORDS:
Clustering, K-Means, Fuzzy C-Means, Vehicle Crashes, Dunn’s Partition Coefficient, Machine Learning
JOURNAL NAME:
Open Access Library Journal,
Vol.12 No.4,
April
8,
2025
ABSTRACT: Clustering is an unsupervised machine learning technique used to organize unlabeled data into groups based on similarity. This paper applies the K-means and Fuzzy C-means clustering algorithms to a vehicle crash dataset in order to explore various patterns in the data. K-means assigns data points to clusters based on the similarity between the data point and the cluster centroids, which results in partitioning the data into distinct clusters. On the other hand, fuzzy C-means clustering allows data points to belong to multiple clusters simultaneously with varying degrees of membership, providing a more diverse representation of the data. Results show that while K-means clustering is simpler and easier to interpret, fuzzy C-means clustering offers more flexibility and can manage situations where data points may have more cluster assignments.