Raymond T. Ng

LOF: identifying density-based local outliers

For many KDD applications, such as detecting criminal activities in E-commerce, finding the rare instances or the outliers, can be more interesting than finding the common patterns. Existing work in outlier detection regards being an outlier as a binary property. In this paper, we contend that for many scenarios, it is more meaningful to assign to each object a degree of being an outlier. This degree is called the local outlier factor (LOF) of an object. It is local in that the degree depends on how isolated the object is with respect to the surrounding neighborhood. We give a detailed formal analysis showing that LOF enjoys many desirable properties. Using real-world datasets, we demonstrate that LOF can be used to find outliers which appear to be meaningful, but can otherwise not be identified with existing approaches. Finally, a careful performance evaluation of our algorithm confirms we show that our approach of finding local outliers can be practical.

CLARANS: a method for clustering objects for spatial data mining

Spatial data mining is the discovery of interesting relationships and characteristics that may exist implicitly in spatial databases. To this end, this paper has three main contributions. First, it proposes a new clustering method called CLARANS, whose aim is to identify spatial structures that may be present in the data. Experimental results indicate that, when compared with existing clustering methods, CLARANS is very efficient and effective. Second, the paper investigates how CLARANS can handle not only point objects, but also polygon objects efficiently. One of the methods considered, called the IR-approximation, is very efficient in clustering convex and nonconvex polygon objects. Third, building on top of CLARANS, the paper develops two spatial data mining algorithms that aim to discover relationships between spatial and nonspatial attributes. Both algorithms can discover knowledge that is difficult to find with existing spatial data mining algorithms.

Exploratory mining and pruning optimizations of constrained associations rules

From the standpoint of supporting human-centered discovery of knowledge, the present-day model of mining association rules suffers from the following serious shortcomings: (i) lack of user exploration and control, (ii) lack of focus, and (iii) rigid notion of relationships. In effect, this model functions as a black-box, admitting little user interaction in between. We propose, in this paper, an architecture that opens up the black-box, and supports constraint-based, human-centered exploratory mining of associations. The foundation of this architecture is a rich set of constraint constructs, including domain, class, and SQL-style aggregate constraints, which enable users to clearly specify what associations are to be mined. We propose constrained association queries as a means of specifying the constraints to be satisfied by the antecedent and consequent of a mined association. In this paper, we mainly focus on the technical challenges in guaranteeing a level of performance that is commensurate with the selectivities of the constraints in an association query. To this end, we introduce and analyze two properties of constraints that are critical to pruning: anti-monotonicity and succinctness. We then develop characterizations of various constraints into four categories, according to these properties. Finally, we describe a mining algorithm called CAP, which achieves a maximized degree of pruning for all categories of constraints. Experimental results indicate that CAP can run much faster, in some cases as much as 80 times, than several basic algorithms. This demonstrates how important the succinctness and anti-monotonicity properties are, in delivering the performance guarantee.

Distance-based outliers: algorithms and applications

Abstract. This paper deals with finding outliers (exceptions) in large, multidimensional datasets. The identification of outliers can lead to the discovery of truly unexpected knowledge in areas such as electronic commerce, credit card fraud, and even the analysis of performance statistics of professional athletes. Existing methods that we have seen for finding outliers can only deal efficiently with two dimensions/attributes of a dataset. In this paper, we study the notion of DB (distance-based) outliers. Specifically, we show that (i) outlier detection can be done efficiently for large datasets, and for k-dimensional datasets with large values of k (e.g.,

A comprehensive analysis of common copy-number variations in the human genome

k \ge 5

Predicting source code changes by mining change history

); and (ii), outlier detection is a meaningful and important knowledge discovery task.First, we present two simple algorithms, both having a complexity of

Probabilistic logic programming

O(k \: N^2)

Indexing spatio-temporal trajectories with Chebyshev polynomials

, k being the dimensionality and N being the number of objects in the dataset. These algorithms readily support datasets with many more than two attributes. Second, we present an optimized cell-based algorithm that has a complexity that is linear with respect to N, but exponential with respect to k. We provide experimental results indicating that this algorithm significantly outperforms the two simple algorithms for

Fate tracing of mature hepatocytes in mouse liver homeostasis and regeneration

k \leq 4

Optimization of constrained frequent set queries with 2-variable constraints

. Third, for datasets that are mainly disk-resident, we present another version of the cell-based algorithm that guarantees at most three passes over a dataset. Again, experimental results show that this algorithm is by far the best for