Mining strong symbiotic patterns hidden in spatial prevalent co-location patterns
Autor: | Lizhen Wang, Junli Lu, Jiasong Zhao, Yuan Fang |
---|---|
Rok vydání: | 2018 |
Předmět: |
Information Systems and Management
Computer science 02 engineering and technology computer.software_genre Management Information Systems Data set Artificial Intelligence 020204 information systems 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Pruning (decision trees) Data mining computer Software |
Zdroj: | Knowledge-Based Systems. 146:190-202 |
ISSN: | 0950-7051 |
DOI: | 10.1016/j.knosys.2018.02.006 |
Popis: | Spatial co-location patterns represent the subsets of spatial features which are frequently located together in a geographic space. Spatial co-location pattern mining has been a research hot in recent years. However, maybe the features in a prevalent co-location pattern further have more interesting relationships such as symbiotic relationships, competitive relationships or causal relationships. This paper mines symbiotic relationships implied in prevalent co-location patterns from dynamic spatial databases. Firstly, after analyzing the existed definition of symbiotic patterns, a criterion of judging strong symbiotic patterns is proposed. Secondly, a novel algorithm to mine strong symbiotic patterns from prevalent co-location patterns is presented, named basic algorithm. Third, for improving the efficiency of the basic algorithm, an improved algorithm which integrates two expensive operations of the basic algorithm into together, and a pruning strategy with two pruning lemmas are presented. The experiments evaluate the effectiveness and efficiency of the proposed algorithms with “real + synthetic” data sets and the results show that strong symbiotic patterns are more concise and actionable compared to traditional prevalent co-location patterns. |
Databáze: | OpenAIRE |
Externí odkaz: |