Existing co-location mining algorithms require a user provided distance threshold at which prevalent patterns are searched. Since spatial interactions, in reality, may happen at different distances, finding the right distance threshold to mine all true patterns is not easy and a single appropriate threshold may not even exist. A standard co-location mining algorithm also requires a prevalence measure threshold to find prevalent patterns. The prevalence measure values of the true co-location patterns occurring at different distances may vary and finding a prevalence measure threshold to mine all true patterns without reporting random patterns is not easy and sometimes not even possible. In this paper, we propose an algorithm to mine true co-location patterns at multiple distances. Our approach is based on a statistical test and does not require thresholds for the prevalence measure and the interaction distance. We evaluate the efficacy of our algorithm using synthetic and real data sets comparing it with the state-of-the-art co-location mining approach.

In spatial domains, interaction between features gives rise to two types of interaction patterns: co-location and segregation patterns. Existing approaches to finding co-location patterns have several shortcomings: (1) They depend on user specified thresholds for prevalence measures; (2) they do not take spatial auto-correlation into account; and (3) they may report co-locations even if the features are randomly distributed. Segregation patterns have yet to receive much attention. In this paper, we propose a method for finding both types of interaction patterns, based on a statistical test. We introduce a new definition of co-location and segregation pattern, we propose a model for the null distribution of features so spatial auto-correlation is taken into account, and we design an algorithm for finding both co-location and segregation patterns. We also develop two strategies to reduce the computational cost compared to a naïve approach based on simulations of the data distribution, and we propose an approach to reduce the runtime of our algorithm even further by using an approximation of the neighborhood of features. We evaluate our method empirically using synthetic and real data sets and demonstrate its advantages over a state-of-the-art co-location mining algorithm.

The increase use of high dimensional, geographically distributed rich and massive meteorological data poses an increasing scientific challenge in efficient outlier mining. Properties in such meteorological data are observed to fluctuate in spatial synchrony. Capturing this spatial variation at different spatial scales requires a multi-resolution analysis. In this paper, we develop an algorithm for region outlier detection at different scales using the multi-resolution feature of wavelet analysis. Another challenge of meteorological data mining is that the data size is huge to accommodate different resolutions and number of samples varies with the spatial scales. This motivated us to design a load adaptive parallel algorithm for outlier detection which can maintain good scalability for all spatial scales. Our algorithm has been implemented on high-performance computing architecture and evaluated on real-world meteorological data.

Summary form only given. Pricing of derivatives is one of the central problems in computational finance. Since the theory of derivative pricing is highly mathematical, numerical techniques such as binomial lattice, finite-differencing and fast Fourier transform (FFT) among others have been used for derivative or option pricing. Based on a recent work on FFT for VLSI circuits, we develop a parallel algorithm in the current work, which improves data locality and hence reduce communication overheads. Our main aim is to study the performance of this algorithm. Compared to the traditional butterfly network, the current algorithm with data swap network performs better by more than 15% for large data sizes.