Unsupervised Instance Selection with Low-Label, Supervised Learning for Outlier Detection
This addresses labeling bottlenecks for practitioners in outlier detection, offering a more efficient alternative to Active Learning, though it is incremental as it builds on existing methods.
The paper tackled the problem of labeling inefficiency in supervised outlier detection by proposing an unsupervised instance selection (UNISEL) technique, which performed comparably to Active Learning and showed superior performance when combined with it on 10 datasets.
The laborious process of labeling data often bottlenecks projects that aim to leverage the power of supervised machine learning. Active Learning (AL) has been established as a technique to ameliorate this condition through an iterative framework that queries a human annotator for labels of instances with the most uncertain class assignment. Via this mechanism, AL produces a binary classifier trained on less labeled data but with little, if any, loss in predictive performance. Despite its advantages, AL can have difficulty with class-imbalanced datasets and results in an inefficient labeling process. To address these drawbacks, we investigate our unsupervised instance selection (UNISEL) technique followed by a Random Forest (RF) classifier on 10 outlier detection datasets under low-label conditions. These results are compared to AL performed on the same datasets. Further, we investigate the combination of UNISEL and AL. Results indicate that UNISEL followed by an RF performs comparably to AL with an RF and that the combination of UNISEL and AL demonstrates superior performance. The practical implications of these findings in terms of time savings and generalizability afforded by UNISEL are discussed.