Narrow Results

A spatial outlier is a spatially referenced object whose non-spatial attribute values are significantly different from the values of its neighborhood. Identification of spatial outliers can lead to the discovery of unexpected, interesting, and useful spatial patterns for further analysis. Previous work in spatial outlier detection focuses on detecting spatial outliers with a single attribute. In the paper, we propose two approaches to discover spatial outliers with multiple attributes. We formulate the multi-attribute spatial outlier detection problem in a general way, provide two effective detection algorithms, and analyze their computation complexity. In addition, using a real-world census data, we demonstrate that our approaches can effectively identify local abnormality in large spatial data sets.

A spatial co-location pattern is a group of spatial objects whose instances are frequently located in the same region. The spatial co-location pattern mining problem has been investigated extensively in the past due to its broad range of applications. In this paper we study this problem for fuzzy objects. Fuzzy objects play an important role in many areas, such as the geographical information system and the biomedical image database. In this paper, we propose two new kinds of co-location pattern mining for fuzzy objects, single co-location pattern mining (SCP) and range co-location pattern mining (RCP), to mining co-location patterns at a membership threshold or within a membership range. For efficient SCP mining, we optimize the basic mining algorithm to accelerate the co-location pattern generation. To improve the performance of RCP mining, effective pruning strategies are developed to significantly reduce the search space. The efficiency of our proposed algorithms as well as the optimization techniques are verified with an extensive set of experiments.

When attempting knowledge discovery on spatial data, certain additional constraints on and relationships among the data must be considered. These include spatially or locationally explicit attributes, as well as more implicit topological relationships. Given such additional constraints, many generalized data mining techniques and algorithms may be specially tailored for mining in spatial data. This chapter introduces several adapted techniques and algorithms that may be applied in a spatial data mining task.

Taxis equipped with GPS can record their trajectory and generate huge amounts of data. We can analyzethe behavior of taxi drivers and search for similarities and common characteristics in their working patterns. In this paper, we utilize taxi GPS data collected form Tianjin city to analyze taxi drivers’ working pattern. Firstly, we determined taxis’ parking place by detecting stopping points and measured the operating range. Secondly, we studied the taxis’ working pattern by comparing the relationship between three representative locations in taxis’ behavior: parking place, working center and city center. Thirdly, we analyzed the spatial pattern from two perspectives of direction and distance. Lastly, we studied the income efficiency of different taxis from different parking places and working centers. The research results demonstrated that the taxis’ individual mobility behavior has clear similarities: The taxi drivers’ operating pattern has a characteristic of moving toward the city center andmost of the working centers distribute in positions between parking places and the city center. The discoveries are significant for urban public infrastructure construction, government’s management of the taxis and taxi drivers’ strategy selections.