This paper introduces Subspace Weighting Clustering (SWC), a novel algorithm designed for analyzing vast stock price datasets. SWC enhances decision-making in the stock market by identifying strong correlations among stocks within specific price-value sub-sequences over time. The algorithm incorporates dimension weights to assess the significance of different dimensions in distinguishing stock clusters, improving the iterative fuzzy k-means clustering procedure. In the experimental results, a synthetic dataset was generated with a known number of clusters and varying cluster distributions to assess the performance of the SWC algorithm. The findings indicate that SWC outperformed the state-of-the-art alternatives on synthetic data and achieves superior returns when applied to real-world stock market data, outperforming a well-known value investment strategy.