The mining of prevalent co-location patterns (PCPs) is a pivotal task in spatial data analysis, providing insights into the co-occurrence relationships among spatial features. Despite its importance, traditional frameworks for co-location pattern mining often suffer from the generation of an exponential number of patterns, many of which are redundant or insignificant. This proliferation of patterns poses significant challenges for practical applications, where the identification of concise and prevalent patterns is essential. In this paper, we delve into the problem of efficient redundancy elimination in PCPs mining. By adopting a post-mining framework, we aim to address the limitations of traditional methods. First, we introduce the novel concept of semantic distance, which measures the relationship between a co-location pattern and its super-patterns, allowing us to discern meaningful patterns from redundant ones. Additionally, we propose the concept of \(\Omega \) -Obscured, which helps in identifying and excluding patterns that are obscured by their more prevalent super-patterns. Our approach is rigorously evaluated through extensive performance studies on both synthetic and real-world datasets. The experimental results demonstrate that our method significantly reduces the number of redundant patterns while preserving the most informative and prevalent co-location patterns. This not only enhances the interpretability of the results but also improves the efficiency of the pattern mining process.