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

Word embeddings improve the performance of NLP systems by revealing thehidden structural relationships between words. Despite their success in manyapplications, word embeddings have seen very little use in computational socialscience NLP tasks, presumably due to their reliance on big data, and to a lackof interpretability. I propose a probabilistic model-based word embeddingmethod which can recover interpretable embeddings, without big data. The keyinsight is to leverage mixed membership modeling, in which globalrepresentations are shared, but individual entities (i.e. dictionary words) arefree to use these representations to uniquely differing degrees. I show how totrain the model using a combination of state-of-the-art training techniques forword embeddings and topic models. The experimental results show an improvementin predictive language modeling of up to 63% in MRR over the skip-gram, anddemonstrate that the representations are beneficial for supervised learning. Iillustrate the interpretability of the models with computational social sciencecase studies on State of the Union addresses and NIPS articles.

Mixed Membership Word Embeddings for Computational Social Science