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 Tools annual

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 monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Machine learning techniques applied to the Natural Language Processing (NLP)component of conversational agent development show promising results forimproved accuracy and quality of feedback that a conversational agent canprovide. The effort required to develop an educational scenario specificconversational agent is time consuming as it requires domain experts to labeland annotate noisy data sources such as classroom videos. Previous approachesto modeling annotations have relied on labeling thousands of examples andcalculating inter-annotator agreement and majority votes in order to model thenecessary scenarios. This method, while proven successful, ignores individualannotator strengths in labeling a data point and under-utilizes examples thatdo not have a majority vote for labeling. We propose using a multi-task weaksupervision method combined with active learning to address these concerns.This approach requires less labeling than traditional methods and showssignificant improvements in precision, efficiency, and time-requirements thanthe majority vote method (Ratner 2019). We demonstrate the validity of thismethod on the Google Jigsaw data set and then propose a scenario to apply thismethod using the Instructional Quality Assessment(IQA) to define the categoriesfor labeling. We propose using probabilistic modeling of annotator labeling togenerate active learning examples to further label the data. Active learning isable to iteratively improve the training performance and accuracy of theoriginal classification model. This approach combines state-of-the art labelingtechniques of weak supervision and active learning to optimize results in theeducational domain and could be further used to lessen the data requirementsfor expanded scenarios within the education domain through transfer learning.

Improving Classification through Weak Supervision in Context-specific Conversational Agent Development for Teacher Education