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

With the advent of strong pre-trained natural language processing models likeBERT, DeBERTa, MiniLM, T5, the data requirement for industries to fine-tunethese models to their niche use cases has drastically reduced (typically to afew hundred annotated samples for achieving a reasonable performance). However,the availability of even a few hundred annotated samples may not always beguaranteed in low resource domains like automotive, which often limits theusage of such deep learning models in an industrial setting. In this paper weaim to address the challenge of fine-tuning such pre-trained models with only afew annotated samples, also known as Few-shot learning. Our experiments focuson evaluating the performance of a diverse set of algorithms and methodologiesto achieve the task of classifying BOSCH automotive domain textual softwarerequirements into 3 categories, while utilizing only 15 annotated samples percategory for fine-tuning. We find that while SciBERT and DeBERTa based modelstend to be the most accurate at 15 training samples, their performanceimprovement scales minimally as the number of annotated samples is increased to50 in comparison to Siamese and T5 based models.

Few-shot learning approaches for classifying low resource domain specific software requirements