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

In recent years, several online platforms have seen a rapid increase in thenumber of review systems that request users to provide aspect-level feedback.Document-level Multi-aspect Sentiment Classification (DMSC), where the goal isto predict the ratings/sentiment from a review at an individual aspect level,has become a challenging and imminent problem. To tackle this challenge, wepropose a deliberate self-attention-based deep neural network model, namelyFEDAR, for the DMSC problem, which can achieve competitive performance whilealso being able to interpret the predictions made. FEDAR is equipped with ahighway word embedding layer to transfer knowledge from pre-trained wordembeddings, an RNN encoder layer with output features enriched by pooling andfactorization techniques, and a deliberate self-attention layer. In addition,we also propose an Attention-driven Keywords Ranking (AKR) method, which canautomatically discover aspect keywords and aspect-level opinion keywords fromthe review corpus based on the attention weights. These keywords aresignificant for rating predictions by FEDAR. Since crowdsourcing annotation canbe an alternate way to recover missing ratings of reviews, we propose aLEcture-AuDience (LEAD) strategy to estimate model uncertainty in the contextof multi-task learning, so that valuable human resources can focus on the mostuncertain predictions. Our extensive set of experiments on five differentopen-domain DMSC datasets demonstrate the superiority of the proposed FEDAR andLEAD models. We further introduce two new DMSC datasets in the healthcaredomain and benchmark different baseline models and our models on them.Attention weights visualization results and visualization of aspect and opinionkeywords demonstrate the interpretability of our model and the effectiveness ofour AKR method.

An Interpretable and Uncertainty Aware Multi-Task Framework for Multi-Aspect Sentiment Analysis