Back to Search
Start Over
A New Effective and Efficient Measure for Outlying Aspect Mining
- Source :
- Web Information Systems Engineering – WISE 2020 ISBN: 9783030620073, WISE (2)
- Publication Year :
- 2020
- Publisher :
- Springer International Publishing, 2020.
-
Abstract
- Outlying Aspect Mining (OAM) aims to find the subspaces (a.k.a. aspects) in which a given query is an outlier with respect to a given data set. Existing OAM algorithms use traditional distance/density-based outlier scores to rank subspaces. Because these distance/density-based scores depend on the dimensionality of subspaces, they cannot be compared directly between subspaces of different dimensionality. Z-score normalisation has been used to make them comparable. It requires to compute outlier scores of all instances in each subspace. This adds significant computational overhead on top of already expensive density estimation—making OAM algorithms infeasible to run in large and/or high-dimensional datasets. We also discover that Z-score normalisation is inappropriate for OAM in some cases. In this paper, we introduce a new score called Simple Isolation score using Nearest Neighbor Ensemble (SiNNE), which is independent of the dimensionality of subspaces. This enables the scores in subspaces with different dimensionalities to be compared directly without any additional normalisation. Our experimental results revealed that SiNNE produces better or at least the same results as existing scores; and it significantly improves the runtime of an existing OAM algorithm based on beam search.
- Subjects :
- Rank (linear algebra)
business.industry
Computer science
Pattern recognition
02 engineering and technology
Linear subspace
k-nearest neighbors algorithm
Data set
ComputingMethodologies_PATTERNRECOGNITION
020204 information systems
Outlier
0202 electrical engineering, electronic engineering, information engineering
Beam search
020201 artificial intelligence & image processing
Artificial intelligence
business
Subspace topology
Curse of dimensionality
Subjects
Details
- ISBN :
- 978-3-030-62007-3
- ISBNs :
- 9783030620073
- Database :
- OpenAIRE
- Journal :
- Web Information Systems Engineering – WISE 2020 ISBN: 9783030620073, WISE (2)
- Accession number :
- edsair.doi...........418975cdc412a90aace6a09d939a75b7