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

In this paper, we study the problem of author identification underdouble-blind review setting, which is to identify potential authors giveninformation of an anonymized paper. Different from existing approaches thatrely heavily on feature engineering, we propose to use network embeddingapproach to address the problem, which can automatically represent nodes intolower dimensional feature vectors. However, there are two major limitations inrecent studies on network embedding: (1) they are usually general-purposeembedding methods, which are independent of the specific tasks; and (2) most ofthese approaches can only deal with homogeneous networks, where theheterogeneity of the network is ignored. Hence, challenges faced here are twofolds: (1) how to embed the network under the guidance of the authoridentification task, and (2) how to select the best type of information due tothe heterogeneity of the network.  To address the challenges, we propose a task-guided and path-augmentedheterogeneous network embedding model. In our model, nodes are first embeddedas vectors in latent feature space. Embeddings are then shared and jointlytrained according to task-specific and network-general objectives. We extendthe existing unsupervised network embedding to incorporate meta paths inheterogeneous networks, and select paths according to the specific task. Theguidance from author identification task for network embedding is provided bothexplicitly in joint training and implicitly during meta path selection. Ourexperiments demonstrate that by using path-augmented network embedding withtask guidance, our model can obtain significantly better accuracy atidentifying the true authors comparing to existing methods.

Task-Guided and Path-Augmented Heterogeneous Network Embedding for Author Identification