Due to the rapid development of wireless communication services and the growth of the Internet of Things, the requirement for spectrum resources is growing significantly. Additionally, due to shadowing, fading, and uncertainty in the receiver, the overall performance of the spectrum sensing technique was compromised. By utilizing the spatial diversity of the nodes, the aforementioned issues can be overcome by following a cooperative spectrum sensing (CSS) approach. This approach slightly increases the overhead and the time consumed for sensing the spectrum. This work introduces a comparative study of clustering methods based on machine learning algorithms using the global positioning system (GPS) coordinates in CSS. We compare five clustering algorithms (K-means, K-Medoids, miniBatchKMeans, Agglomerative, Affinity Propagation), the first four of which require a K value to be specified. These algorithms require an extra step compared to the Affinity Propagation algorithm. In this work, we propose CSS-based clustering ML algorithms and a technique for updating the used model.