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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

State-of-the-art weakly supervised text classification methods, whilesignificantly reduced the required human supervision, still requires thesupervision to cover all the classes of interest. This is never easy to meet inpractice when human explore new, large corpora without complete pictures. Inthis paper, we work on a novel yet important problem of weakly supervisedopen-world text classification, where supervision is only needed for a fewexamples from a few known classes and the machine should handle both known andunknown classes in test time. General open-world classification has beenstudied mostly using image classification; however, existing methods typicallyassume the availability of sufficient known-class supervision and strongunknown-class prior knowledge (e.g., the number and/or data distribution). Wepropose a novel framework WOT-Class that lifts those strong assumptions.Specifically, it follows an iterative process of (a) clustering text to newclasses, (b) mining and ranking indicative words for each class, and (c)merging redundant classes by using the overlapped indicative words as a bridge.Extensive experiments on 7 popular text classification datasets demonstratethat WOT-Class outperforms strong baselines consistently with a large margin,attaining 23.33% greater average absolute macro-F1 over existing approachesacross all datasets. Such competent accuracy illuminates the practicalpotential of further reducing human effort for text classification.

WOT-Class: Weakly Supervised Open-world Text Classification