Computational Intelligence Approach for Exploration of Spatial Co‐Location Patterns
S. LourduMarie Sophie, S. Siva Sathya, S. Sharmiladevi, J. Dhakshayani · 2024
A spatial co-location pattern is a collection of spatial occurrences as they tend to appear in contiguous geographic vicinity. Numerous co-location algorithms with the objective of identifying lossless condensed spatial co-location patterns have been presented, but these methods are sequential. Co-location patterns are subgroups of spatial characteristics whose instances are regularly situated in close proximity. The issue is crucial in numerous applications, like the analysis of crime or illness associations with environmental parameters; however, it is computationally difficult given the huge number of occurrences. This work proposes a computational intelligence approach for detecting unique spatial patterns in a clustered neighborhood by employing a Grid Clustered technique. The Top-K co-location technique is also utilized to generate the most highly co-located spatial patterns. The suggested technique is implemented using the MapReduce framework, a parallel processing framework that enables rapid and efficient processing. MapReduce increases the efficiency of spatial pattern mining and works best with big spatial datasets. The experimental assessment offers a summary of the algorithm's efficacy across a spectrum of data sizes.