Abstract: Distanccbascd outlicr detection approach typically requires time of distance computation and comparison. This quadratic scaling restricts the ability to apply this approach to large datasets. To overcome this limitation, a novel distance-based outlier mining approach with pruning rules was proposed. The approach consists of two phases.During the first phase, the original input data arc scanned and the majority of non-outlicrs arc pruned. During second phase, an improved nested loops approach is applied to compute the average K-nearest distance which measures the degree of being an outlicr and finally reports the top-n outlicrs. Experiments on both synthetic data and real-life data show hat the proposed approach achieves a high hit rate with a low false alarm rate. Compared with related approaches, theproposed approach has a lower time complexity.

Key words: Outlicr, Data mining, Distanccbascd