Abstract: In the current scenario, a vast amount of unlabeled high-dimensional data exhibits intrinsic relationships, making it suitable for information extraction through graph-based clustering ...