English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Support Vector Data Description (SVDD) is a popular one-class classifiers foranomaly and novelty detection. But despite its effectiveness, SVDD does notscale well with data size. To avoid prohibitive training times, samplingmethods select small subsets of the training data on which SVDD trains adecision boundary hopefully equivalent to the one obtained on the full dataset. According to the literature, a good sample should therefore containso-called boundary observations that SVDD would select as support vectors onthe full data set. However, non-boundary observations also are essential to notfragment contiguous inlier regions and avoid poor classification accuracy.Other aspects, such as selecting a sufficiently representative sample, areimportant as well. But existing sampling methods largely overlook them,resulting in poor classification accuracy. In this article, we study how toselect a sample considering these points. Our approach is to frame SVDDsampling as an optimization problem, where constraints guarantee that samplingindeed approximates the original decision boundary. We then propose RAPID, anefficient algorithm to solve this optimization problem. RAPID does not requireany tuning of parameters, is easy to implement and scales well to large datasets. We evaluate our approach on real-world and synthetic data. Our evaluationis the most comprehensive one for SVDD sampling so far. Our results show thatRAPID outperforms its competitors in classification accuracy, in sample size,and in runtime.

Efficient SVDD Sampling with Approximation Guarantees for the Decision Boundary