Improvement of Fuzzy Geographically Weighted Clustering-Ant Colony Optimization Performance using Context-Based Clustering and CUDA Parallel Programming

Abstract

Abstract. Geo-demographic analysis (GDA) is the study of population characteristics by geographical area. Fuzzy Geographically Weighted Clustering (FGWC) is an effective algorithm used in GDA. Improvement of FGWC has been done by integrating a metaheuristic algorithm, Ant Colony Optimization (ACO), as a global optimization tool to increase the clustering accuracy in the initial stage of the FGWC algorithm. However, using ACO in FGWC increases the time to run the algorithm compared to the standard FGWC algorithm. In this paper, context-based clustering and CUDA parallel programming are proposed to improve the performance of the improved algorithm (FGWC-ACO). Contextbased clustering is a method that focuses on the grouping of data based on certain conditions, while CUDA parallel programming is a method that uses the graphical processing unit (GPU) as a parallel processing tool. The Indonesian Population Census 2010 was used as the experimental dataset. It was shown that the proposed methods were able to improve the performance of FGWC-ACO without reducing the clustering quality of the original method. The clustering quality was evaluated using the clustering validity index.

Micro, Small, and Medium Enterprises’ Business Vulnerability Cluster in Indonesia: An Analysis Using Optimized Fuzzy Geodemographic Clustering

“…This is consistent with the previous studies that found that the modification of FGWC can lead to a better clustering quality. The modification is not just limited to the metaheuristic optimization but also the modification of the spatial interaction [32], distance matrix [33], or the hybridization of context-based and fast computing like CUDA [34]. On the other hand, comparing the optimization algorithms, the FPA has the best performance in optimizing the FGWC clustering result in the business vulnerability context.…”

“…This study also used a switching probability of 0.7, which is close to Yang [37], who stated that 0.8 works better in several studies. Future studies may improve FPA optimization using some modifications, such as multiobjective FPA [69], distance steps setting using other distributions as in [70], implementing CUDA for increasing speed [34], etc.…”

Fuzzy Geographically Weighted Clustering dengan Gravitational Search Algorithm pada Kasus Penyandang Masalah Kesejahteraan Sosial di Provinsi Jawa Tengah

“…Namun FGWC mudah terjebak ke dalam lokal optima. Hal tersebut disebabkan inisialisasi pusat awal klaster dilakukan secara acak (Nurmala & Purwarianti, 2017). Untuk menangani keterbatasan tersebut digunakan Gravitational Search Algorithm (GSA).…”