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

In large-scale domain classification, an utterance can be handled by multipledomains with overlapped capabilities. However, only a limited number ofground-truth domains are provided for each training utterance in practice whileknowing as many as correct target labels is helpful for improving the modelperformance. In this paper, given one ground-truth domain for each trainingutterance, we regard domains consistently predicted with the highestconfidences as additional pseudo labels for the training. In order to reduceprediction errors due to incorrect pseudo labels, we leverage utterances withnegative system responses to decrease the confidences of the incorrectlypredicted domains. Evaluating on user utterances from an intelligentconversational system, we show that the proposed approach significantlyimproves the performance of domain classification with hypothesis reranking.

Pseudo Labeling and Negative Feedback Learning for Large-scale Multi-label Domain Classification