Sammenlign metoder
Gennemgå dine valgte metoder side om side; rækker, der afviger, er fremhævet.
| Aktiv læring One-class SVM× | Semi-supervised Learning× | |
|---|---|---|
| Fagområde | Maskinlæring | Maskinlæring |
| Familie | Machine learning | Machine learning |
| Oprindelsesår≠ | 2000s | 1970s–2006 (formalized) |
| Ophavsperson≠ | Schölkopf et al. (OCSVM); active variant developed in the anomaly-detection literature (2000s–2010s) | Vapnik, V. N. and others (community of researchers, 1970s–2000s) |
| Type≠ | Semi-supervised anomaly/novelty detection with iterative labeling | Learning paradigm |
| Oprindelig kilde≠ | Schölkopf, B., Platt, J. C., Shawe-Taylor, J., Smola, A. J., & Williamson, R. C. (1999). Estimating the Support of a High-Dimensional Distribution. Neural Computation, 13(7), 1443–1471. DOI ↗ | Chapelle, O., Scholkopf, B., & Zien, A. (Eds.) (2006). Semi-Supervised Learning. MIT Press. ISBN: 978-0-262-03358-9 |
| Aliasser | AL-OCSVM, active one-class SVM, active novelty detection SVM, query-driven OCSVM | SSL, semi-supervised machine learning, transductive learning, label-efficient learning |
| Relaterede≠ | 4 | 5 |
| Resumé≠ | Active Learning One-class SVM combines the one-class support vector machine — a kernel-based novelty detector that learns the boundary of normal data — with an active learning loop that selects the most informative unlabeled instances for expert annotation. The result is a data-efficient anomaly detector that improves its decision boundary with minimal labeling effort. | Semi-supervised learning (SSL) is a machine learning paradigm that trains models using a small set of labeled examples together with a much larger pool of unlabeled data. By leveraging the structure inherent in unlabeled data, SSL achieves accuracy closer to fully supervised models while requiring far fewer costly manual labels — making it practical when labeling is expensive, slow, or resource-constrained. |
| ScholarGateDatasæt ↗ |
|
|