Spatial Co-locations Research Articles

Spatial co-location pattern mining over extended objects based on cell-relation operations

Spatial co-location pattern mining (SCPM) is intended to discover subsets of spatial features whose instances are frequently located together in geographic areas. Traditional SCPM methods are designed for point spatial instances. However, in reality, instances are mostly in the form of extended objects, e.g., lines, polygons. In addition, current SCPM methods with extended objects are less well researched and have two disadvantages: (1) Existing researches cannot effectively capture neighborhood relationships between extended objects and their mining results cannot properly reflect the distribution dependence of features; (2) These methods are not efficient enough with large datasets. This paper proposes a novel framework called cell-relation operations framework to overcome these issues. To eliminate the first shortcoming, the framework uses the area overlapping of buffers between objects to gain the neighbor relationships between extended objects and introduces participation index under buffer size k to identify prevalent co-location patterns. To address the second problem, our framework employs cell-relation operations rather than instance relation computing as the basic computing unit for co-location mining, which substantially speeds up the computation. The framework obtains spatial co-locations by counting the feature transactions of the cells and calculates the feature overlap ratio of the cells to generate co-locations. We implement experiments with real datasets to demonstrate that our framework’s mining results are more reasonable and the proposed framework’s runtime outperforms the baselines by 2 to 4 orders of magnitude.

Discovering regions of anomalous spatial co-locations

ABSTRACT Regions of anomalous spatial co-locations (ROASCs) are regions where co-locations between two different features are significantly stronger or weaker than expected. ROASC discovery can provide useful insights for studying unexpected spatial associations at regional scales. The main challenges are that the ROASCs are spatially arbitrary in geographic shape and the distributions of spatial features are unknown a priori. To avoid restrictive assumptions regarding the distribution of data, we propose a distribution-free method for discovering arbitrarily shaped ROASCs. First, we present a multidirectional optimization method to adaptively identify the candidate ROASCs, whose sizes and shapes are fully endogenized. Furthermore, the validity of the candidates is evaluated through significance tests under the null hypothesis that the expected spatial co-locations between two features occur consistently across space. To effectively model the null hypothesis, we develop a bivariate pattern reconstruction method by reconstructing the spatial auto- and cross-correlation structures observed in the data. Synthetic experiments and a case study conducted using Shanghai taxi datasets demonstrate the advantages of our method, in terms of effectiveness, over an available alternative method.

Mining spatial colocations from image-objects: A tensor factorization approach

Mining spatial colocations from image-objects: A tensor factorization approach

MINING CO-LOCATION PATTERNS WITH CLUSTERING ITEMS FROM SPATIAL DATA SETS

Abstract. The explosive growth of spatial data and widespread use of spatial databases emphasize the need for the spatial data mining. Co-location patterns discovery is an important branch in spatial data mining. Spatial co-locations represent the subsets of features which are frequently located together in geographic space. However, the appearance of a spatial feature C is often not determined by a single spatial feature A or B but by the two spatial features A and B, that is to say where A and B appear together, C often appears. We note that this co-location pattern is different from the traditional co-location pattern. Thus, this paper presents a new concept called clustering terms, and this co-location pattern is called co-location patterns with clustering items. And the traditional algorithm cannot mine this co-location pattern, so we introduce the related concept in detail and propose a novel algorithm. This algorithm is extended by join-based approach proposed by Huang. Finally, we evaluate the performance of this algorithm.

Effective lossless condensed representation and discovery of spatial co-location patterns

A spatial co-location pattern is a set of spatial features frequently co-occuring in nearby geographic spaces. Similar to closed frequent itemset mining, closed co-location pattern (CCP) mining was proposed for losslessly condensing large collections of prevalent co-location patterns. However, the state-of-the-art condensation methods in mining CCP are inspired by closed frequent itemset mining and do not consider the intrinsic characteristics of spatial co-locations, e.g., the participation index and ratio in spatial feature interactions, thus causing serious containment issues in CCP mining. In this paper, we propose a novel lossless condensed representation of prevalent co-location patterns, Super Participation Index-closed (SPI-closed) co-location. An efficient SPI-closed Miner is also proposed to effectively capture the nature of spatial co-location patterns, alongside the development of three additional pruning strategies to make the SPI-closed Miner efficient. This method captures richer feature interactions in spatial co-locations and solves the containment issues in existing CCP methods. A performance evaluation conducted on both synthetic and real-life data sets shows that SPI-closed Miner reduces the number of CCPs by up to 50%, and runs much faster than the baseline CCP mining algorithm described in the literature.

空间并置模式挖掘研究

空间并置模式挖掘研究

Parallel Approach for Finding Co-location Pattern – A Map Reduce Framework

Spatial co-location pattern mining is a sub field of data mining which is used to discover interesting patterns which are expressed as co-location rules. The objects that are frequently located in certain region are expressed as spatial co-locations. It presents a challenge for finding co-location patterns as the traditional data is considered discrete whereas the spatial objects are embedded in a continuous space. For this a join-less approach is proposed, but as the data size increases, a large amount of computation time is devoted to find co-location rules as the approach is purely sequential. We propose a parallelized join-less approach which finds the spatial neighbor relationship in order to identify co-location instances and co-location rules. The proposed work decreases the computation time drastically as it uses a Map-Reduce framework. This paper presents precise and completeness of the new approach. Finally, an experimental evaluations using synthetic data sets show the algorithm is computationally more efficient.

Spatial co-location pattern discovery without thresholds

Spatial co-location pattern mining discovers the subsets of features whose events are frequently located together in geographic space. The current research on this topic adopts a threshold-based approach that requires users to specify in advance the thresholds of distance and prevalence. However, in practice, it is not easy to specify suitable thresholds. In this article, we propose a novel iterative mining framework that discovers spatial co-location patterns without predefined thresholds. With the absolute and relative prevalence of spatial co-locations, our method allows users to iteratively select informative edges to construct the neighborhood relationship graph until every significant co-location has enough confidence and eventually to discover all spatial co-location patterns. The experimental results on real world data sets indicate that our framework is effective for prevalent co-locations discovery.

An order-clique-based approach for mining maximal co-locations

Most algorithms for mining spatial co-locations adopt an Apriori-like approach to generate size- k prevalence co-locations after size-( k − 1) prevalence co-locations. However, generating and storing the co-locations and table instances is costly. A novel order-clique-based approach for mining maximal co-locations is proposed in this paper. The efficiency of the approach is achieved by two techniques: (1) the spatial neighbor relationships and the size-2 prevalence co-locations are compressed into extended prefix-tree structures, which allows the order-clique-based approach to mine candidate maximal co-locations and co-location instances; and (2) the co-location instances do not need to be stored after computing some characteristics of the corresponding co-location, which significantly reduces the execution time and space required for mining maximal co-locations. The performance study shows that the new method is efficient for mining both long and short co-location patterns, and is faster than some other methods (in particular the join-based method and the join-less method).

A Joinless Approach for Mining Spatial Colocation Patterns

Spatial colocations represent the subsets of features which are frequently located together in geographic space. Colocation pattern discovery presents challenges since spatial objects are embedded in a continuous space, whereas classical data is often discrete. A large fraction of the computation time is devoted to identifying the instances of colocation patterns. We propose a novel joinless approach for efficient colocation pattern mining. The joinless colocation mining algorithm uses an instance-lookup scheme instead of an expensive spatial or an instance join operation for identifying colocation instances. We prove the joinless algorithm is correct and complete in finding colocation rules. We also describe a partial join approach for a spatial data set often clustered in neighborhood areas. We provide the algebraic cost models to characterize the performance dominance zones of the joinless method and the partial join method with a current join-based colocation mining method, and compare their computational complexities. In the experimental evaluation, using synthetic and real-world data sets, our methods performed more efficiently than the join-based method and show more scalability in dense data.