Fuzzy Set Theory-Based Approach for Mining Spatial Association Rules: Road Accident as a Case Study

Abstract

AbstractDespite intensive research over the last decades in spatial data mining, a considerable amount of collected data brings new challenges to the extraction of spatial association rules. Notwithstanding, association rules are examining the numerical and categorical data accurately; however, it is not directly extensible to efficiently analyze real-world spatial data. Analyzing spatial data exposes significant difficulties, including particular representations of spatial information like geometrical measures, topological measures, and spatial object dependencies. Most current studies ignore the integration of geographical information in the association rules mining process, which is not beneficial for the decision-makers to extract relevant association rules. To deal with this issue, we proposed a fuzzy set theory-based approach for mining spatial association rules; This proposed approach would be especially valuable for decision-makers suffering from imprecision and non-quality of extracted association rules. An analysis of road accidents using spatial data shows that the use of fuzzy set theory improves the quality and the precision of extracted association rules. Overall, the proposed approach contributes to a better understanding of spatial association rules and provides meaningful information that can aid decision-makers to improve the performances and precision of spatial association rules.KeywordsData miningSpatial association rulesSpatial predicatesFuzzy logicRoad accidentsGeographical data