Anomaly Detection for Data with Spatial Attributes

Abstract

The problem of detecting spatially-coherent groups of data that exhibit anomalous behavior has started to attract attention due to applications across areas such as epidemic analysis and weather forecasting. Earlier efforts from the data mining community have largely focused on finding outliers, individual data objects that display deviant behavior. Such point-based methods are not easy to extend to find groups of data that exhibit anomalous behavior. Scan statistics are methods from the statistics community that have considered the problem of identifying regions where data objects exhibit a behavior that is atypical of the general dataset. The spatial scan statistic and methods that build upon it mostly adopt the framework of defining a character for regions (e.g., circular or elliptical) of objects and repeatedly sampling regions of such character followed by applying a statistical test for anomaly detection. In the past decade, there have been efforts from the statistics community to enhance efficiency of scan statistics as well as to enable discovery of arbitrarily shaped anomalous regions. On the other hand, the data mining community has started to look at determining anomalous regions that have behavior divergent from their neighborhood. In this chapter, we survey the space of techniques for detecting anomalous regions on spatial data from across the data mining and statistics communities while outlining connections to wellstudied problems in clustering and image segmentation. We analyze the techniques systematically by categorizing them appropriately to provide a structured birds-eye view of the work on anomalous region detection; we hope that this would encourage better cross-pollination of ideas across communities to help advance the frontier in anomaly detection.

Machine Learning: A New Prospect in Multi-Omics Data Analysis of Cancer

“…Clustering, association, and dimensionality reduction are among the techniques used in this approach (Ghahramani, 2003;Hastie et al, 2009). In clustering technique, the main idea is zoomed in finding the pattern with intention to divide the data into several groups with common attributes (Deepak, 2016). This is while, the association technique deals with discovering relationship among variables which can have a great contribution to discover knowledge from a data set (Cios et al, 2007).…”

An End-to-End Computer Vision Methodology for Quantitative Metallography

“…As previously explained, an important application for anomaly detection in the context of materials sciences and others, is detecting defects. These defects normally span an anomalous area of several objects, rather than just a single anomalous object 83 . For this reason we present a novel clustering algorithm, which we call Market Clustering, that divides impurities into anomalous areas, based on the anomaly scores of the impurities from the previous anomaly measures.…”

Detecting statistically significant geographical anomalous regions from spatial sampling points by coupling Gaussian function and multidirectional optimization

“…A spatial neighborhood is typically defined as a set of objects with spatial proximity to the object under consideration (Deepak, 2016). For gridded or polygon datasets, the spatial neighborhoods can be determined directly based on the topological relationships between spatial objects (Han, Kamber, & Pei, 2012).…”

“…Most existing methods have certain limitations in accurately depicting the outlines of various anomalous regions, such as the sensitivity to multiple parameters. Additionally, the significance testing of anomalous regions should also be taken into account to guarantee their validity (Deepak, 2016).…”