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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

In this paper, we present a novel method for detecting fake and LargeLanguage Model (LLM)-generated profiles in the LinkedIn Online Social Networkimmediately upon registration and before establishing connections. Early fakeprofile identification is crucial to maintaining the platform's integrity sinceit prevents imposters from acquiring the private and sensitive information oflegitimate users and from gaining an opportunity to increase their credibilityfor future phishing and scamming activities. This work uses textual informationprovided in LinkedIn profiles and introduces the Section and Subsection TagEmbedding (SSTE) method to enhance the discriminative characteristics of thesedata for distinguishing between legitimate profiles and those created byimposters manually or by using an LLM. Additionally, the dearth of a largepublicly available LinkedIn dataset motivated us to collect 3600 LinkedInprofiles for our research. We will release our dataset publicly for researchpurposes. This is, to the best of our knowledge, the first large publiclyavailable LinkedIn dataset for fake LinkedIn account detection. Within ourparadigm, we assess static and contextualized word embeddings, including GloVe,Flair, BERT, and RoBERTa. We show that the suggested method can distinguishbetween legitimate and fake profiles with an accuracy of about 95% across allword embeddings. In addition, we show that SSTE has a promising accuracy foridentifying LLM-generated profiles, despite the fact that no LLM-generatedprofiles were employed during the training phase, and can achieve an accuracyof approximately 90% when only 20 LLM-generated profiles are added to thetraining set. It is a significant finding since the proliferation of severalLLMs in the near future makes it extremely challenging to design a singlesystem that can identify profiles created with various LLMs.

The Looming Threat of Fake and LLM-generated LinkedIn Profiles: Challenges and Opportunities for Detection and Prevention