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/

Zendy Pro

Presents the zaia usage as premium feature

ZAIA

Presents AI Discovery search usage as a premium feature

AI Discovery

Presents AI Discovery explain and summary usage as a premium feature

Content Analysis

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Presents the pdf analysis as premium feature

PDF Analysis

1500 Top-up

3000 Top-up

Zendy Basic

2500 Top-up

Zendy Unlimited

Presents the access of premium content as premium feature

Premium Content

Internet censorship is a phenomenon of societal importance and attractsinvestigation from multiple disciplines. Several research groups, such asCensored Planet, have deployed large scale Internet measurement platforms tocollect network reachability data. However, existing studies generally rely onmanually designed rules (i.e., using censorship fingerprints) to detectnetwork-based Internet censorship from the data. While this rule-based approachyields a high true positive detection rate, it suffers from several challenges:it requires human expertise, is laborious, and cannot detect any censorship notcaptured by the rules. Seeking to overcome these challenges, we design andevaluate a classification model based on latent feature representation learningand an image-based classification model to detect network-based Internetcensorship.  To infer latent feature representations fromnetwork reachability data, wepropose a sequence-to-sequence autoencoder to capture the structure and theorder of data elements in the data. To estimate the probability of censorshipevents from the inferred latent features, we rely on a densely connectedmulti-layer neural network model.  Our image-based classification model encodes a network reachability datarecord as a gray-scale image and classifies the image as censored or not usinga dense convolutional neural network. We compare and evaluate both approachesusing data sets from Censored Planet via a hold-out evaluation. Bothclassification models are capable of detecting network-based Internetcensorship as we were able to identify instances of censorship not detected bythe known fingerprints. Latent feature representations likely encode morenuances in the data since the latent feature learning approach discovers agreater quantity, and a more diverse set, of new censorship instances.

Detecting Network-based Internet Censorship via Latent Feature Representation Learning