Discovery of spatial association rules in geo-referenced census data: A relational mining approach

Appice, Annalisa; Ceci, Michelangelo; Lanza, Antonietta; Lisi, Francesca A.; Malerba, Donato

doi:10.3233/ida-2003-7604

Cited by 68 publications

(31 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Common conceptualization of spatial relationship is defined using topological relationships (Ceci et al 2004;Papadias and Theodoridis 1997;Santos and Amaral 2005), a fixed distance (Shekhar and Huang 2001;Shekhar 2004, 2006;Xiao et al 2008), buffer zones (Appice et al 2003;Chelghoum and Zeitouni 2004;Xiong et al 2004), k nearest neighbors (Wan and Zhou 2008), space partition (Bembenik and Rybinski 2008;Frank et al 2009) and so on. Topological relationships, preserved under translation, rotation or scaling (Easter 1991), define the relationship between two objects in terms of eight predicates that describe the intersections of their boundary, interior and exterior (e.g.…”

Section: Neighborhood Definitionsmentioning

confidence: 99%

Mining spatial colocation patterns: a different framework

Yoo

Bow

2011

Data Min Knowl Disc

View full text Add to dashboard Cite

Recently, there has been considerable interest in mining spatial colocation patterns from large spatial datasets. Spatial colocation patterns represent the subsets of spatial events whose instances are often located in close geographic proximity. Most studies of spatial colocation mining require the specification of two parameter constraints to find interesting colocation patterns. One is a minimum prevalent threshold of colocations, and the other is a distance threshold to define spatial neighborhood. However, it is difficult for users to decide appropriate threshold values without prior knowledge of their task-specific spatial data. In this paper, we propose a different framework for spatial colocation pattern mining. To remove the first constraint, we propose the problem of finding N -most prevalent colocated event sets, where N is the desired number of colocated event sets with the highest interest measure values per each pattern size. We developed two alternative algorithms for mining the N -most patterns. They reduce candidate events effectively and use a filter-and-refine strategy for efficiently finding colocation instances from a spatial dataset. We prove the algorithms are correct and complete in finding the N -most prevalent colocation patterns. For the second constraint, a distance threshold for spatial neighborhood determination, we present various methods to estimate appropriate distance bounds from user input data. The result can help an user to set a distance for a conceptualization of spatial neighborhood. Our experimental results with real and synthetic datasets show that our algorithmic design is computationally effective in finding the N -most prevalent colocation patterns. The discovered patterns were different depending on the distance threshold, which shows that it is important to select appropriate neighbor distances.Responsible editor: M.J. Zaki.

show abstract

Section: Neighborhood Definitionsmentioning

confidence: 99%

Mining spatial colocation patterns: a different framework

Yoo

Bow

2011

Data Min Knowl Disc

View full text Add to dashboard Cite

show abstract

“…By assigning a pattern P with an existentially quantified conjunctive formula eqc(P), obtained by transforming P into a Datalog query, the units of analysis D h ½s are covered by a pattern P if D h ½s F eqcðPÞ, namely D h ½s logically entails eqc(P). Frequent patterns are mined with SPADA [26,27], which enables the discovery of relational patterns (at different levels of granularity) whose support exceeds minSup. SPADA performs a breadth-first search of the space of patterns, from the most general to the more specific ones, and prunes portions of the space which contain only non-frequent patterns.…”

Section: Discovering Frequent Relational Patternsmentioning

confidence: 99%

Relational mining for discovering changes in evolving networks

2015

Self Cite

View full text Add to dashboard Cite

“…The other approach is based on qualitative reasoning [2,4,7,11,16,18] and deals with any spatial relationship (e.g. distance, topology, order) between a reference geographic object type and a set of relevant feature types represented by any geometric primitive (e.g.…”

Section: Related Work and Contributionmentioning

confidence: 99%

“…Appice [2] proposed a method to extract all spatial features and spatial relationships to a deductive relational database. This process is computationally expensive since all spatial relationships are computed a priori.…”

Section: Related Work and Contributionmentioning

confidence: 99%

“…However, it is well known that frequent pattern mining often generates a large number of frequent itemsets and rules [20], among which a small number is novel and interesting to the user. In spatial data mining, this problem increases because of the natural dependences among geographic data, which generate a large amount of patterns and associations that are well known [2], [4]. While in transactional databases items are supposed to be independent from each other (e.g.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Semantic-based pruning of redundant and uninteresting frequent geographic patterns

2009

View full text Add to dashboard Cite

In geographic association rule mining many patterns are either redundant or contain well known geographic domain associations explicitly represented in knowledge resources such as geographic database schemas and geo-ontologies. Existing spatial association rule mining algorithms are Apriori-like, and therefore generate a large amount of redundant patterns. For non-spatial data, the closed frequent pattern mining technique has been introduced to remove redundant patterns. This approach, however, does not warrant the elimination of both redundant and well known geographic dependences when mining geographic databases. This paper presents a novel method for pruning both redundant and well known geographic dependences, by pushing semantics into the pattern mining task. Experiments with real geographic databases have demonstrated a significant reduction of the total amount of patterns and the efficiency of the method.

show abstract

Discovery of spatial association rules in geo-referenced census data: A relational mining approach

Cited by 68 publications

References 20 publications

Mining spatial colocation patterns: a different framework

Mining spatial colocation patterns: a different framework

Relational mining for discovering changes in evolving networks

Semantic-based pruning of redundant and uninteresting frequent geographic patterns

Contact Info

Product

Resources

About