LOF: identifying density-based local outliers
SIGMOD '00 Proceedings of the 2000 ACM SIGMOD international conference on Management of data
Algorithms for Mining Distance-Based Outliers in Large Datasets
VLDB '98 Proceedings of the 24rd International Conference on Very Large Data Bases
Feature bagging for outlier detection
Proceedings of the eleventh ACM SIGKDD international conference on Knowledge discovery in data mining
Clustering Using a Similarity Measure Based on Shared Near Neighbors
IEEE Transactions on Computers
A Fast Feature-Based Method to Detect Unusual Patterns in Multidimensional Datasets
DaWaK '09 Proceedings of the 11th International Conference on Data Warehousing and Knowledge Discovery
Hi-index | 0.00 |
In this paper, we present an alternative approach to discover interesting unusual observations that can not be discovered by outlier detection techniques. The unusual pattern is determined according to the deviation of a group of observations from other observations and the number of observations in the group. To measure the degree of deviation, we introduce the concept of adaptive nearest neighbors that captures the natural similarity between two observations. The boundary points determined by the adaptive nearest neighbor algorithm are used to adjust the level of granularity. The adaptive nearest neighbors are then used to cluster the data set. Finally, we ran experiments on a real life data set to evaluate the result. According to the experiments, we discovered interesting unusual patterns that are overlooked by using outlier detection and clustering algorithms.