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

Detection of vaccine adverse events is crucial to the discovery andimprovement of problematic vaccines. To achieve it, traditionally formalreporting systems like VAERS support accurate but delayed surveillance, whilerecently social media have been mined for timely but noisy observations.Utilizing the complementary strengths of these two domains to boost thedetection performance looks good but cannot be effectively achieved by existingmethods due to significant differences between their data characteristics,including: 1) formal language v.s. informal language, 2) single-message peruser v.s. multi-messages per user, and 3) one class v.s. binary class. In thispaper, we propose a novel generic framework named Multi-instance DomainAdaptation (MIDA) to maximize the synergy between these two domains in thevaccine adverse event detection task for social media users. Specifically, wepropose a generalized Maximum Mean Discrepancy (MMD) criterion to measure thesemantic distances between the heterogeneous messages from these two domains intheir shared latent semantic space. Then these message-level generalized MMDdistances are synthesized by newly proposed mixed instance kernels touser-level distances. We finally minimize the distances between the samples ofthe partially-matched classes from these two domains. In order to solve thenon-convex optimization problem, an efficient Alternating Direction Method ofMultipliers (ADMM) based algorithm combined with the Convex-Concave Procedure(CCP) is developed to optimize parameters accurately. Extensive experimentsdemonstrated that our model outperformed the baselines by a large margin undersix metrics. Case studies showed that formal reports and extractedadverse-relevant tweets by MIDA shared a similarity of keyword and descriptionpatterns.

Multi-instance Domain Adaptation for Vaccine Adverse Event Detection