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

Detecting latent structure within a dataset is a crucial step in performinganalysis of a dataset. However, existing state-of-the-art techniques forsubclass discovery are limited: either they are limited to detecting very smallnumbers of outliers or they lack the statistical power to deal with complexdata such as image or audio. This paper proposes a solution to this subclassdiscovery problem: by leveraging instance explanation methods, an existingclassifier can be extended to detect latent classes via differences in theclassifier's internal decisions about each instance. This works not only withsimple classification techniques but also with deep neural networks, allowingfor a powerful and flexible approach to detecting latent structure withindatasets. Effectively, this represents a projection of the dataset into theclassifier's "explanation space," and preliminary results show that thistechnique outperforms the baseline for the detection of latent classes evenwith limited processing. This paper also contains a pipeline for analyzingclassifiers automatically, and a web application for interactively exploringthe results from this technique.

Class Introspection: A Novel Technique for Detecting Unlabeled Subclasses by Leveraging Classifier Explainability Methods