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 Plus

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

Zendy Tools

Zendy Open

Counterfeit apps impersonate existing popular apps in attempts to misguideusers to install them for various reasons such as collecting personalinformation or spreading malware. Many counterfeits can be identified onceinstalled, however even a tech-savvy user may struggle to detect them beforeinstallation. To this end, this paper proposes to leverage the recent advancesin deep learning methods to create image and text embeddings so thatcounterfeit apps can be efficiently identified when they are submitted forpublication. We show that a novel approach of combining content embeddings andstyle embeddings outperforms the baseline methods for image similarity such asSIFT, SURF, and various image hashing methods. We first evaluate theperformance of the proposed method on two well-known datasets for evaluatingimage similarity methods and show that content, style, and combined embeddingsincrease precision@k and recall@k by 10%-15% and 12%-25%, respectively whenretrieving five nearest neighbours. Second, specifically for the appcounterfeit detection problem, combined content and style embeddings achieve12% and 14% increase in precision@k and recall@k, respectively compared to thebaseline methods. Third, we present an analysis of approximately 1.2 millionapps from Google Play Store and identify a set of potential counterfeits fortop-10,000 popular apps. Under a conservative assumption, we were able to find2,040 potential counterfeits that contain malware in a set of 49,608 apps thatshowed high similarity to one of the top-10,000 popular apps in Google PlayStore. We also find 1,565 potential counterfeits asking for at least fiveadditional dangerous permissions than the original app and 1,407 potentialcounterfeits having at least five extra third party advertisement libraries.

A Multi-modal Neural Embeddings Approach for Detecting Mobile Counterfeit Apps: A Case Study on Google Play Store