COIP—Continuous, Operable, Impartial, and Privacy-Aware Identity Validity Estimation for OSN Profiles

Abstract

Identity validation of Online Social Networks’ (OSNs’) peers is a critical concern to the insurance of safe and secure online socializing environments. Starting from the vision of empowering users to determine the validity of OSN identities, we suggest a framework to estimate the trustworthiness of online social profiles based only on the information they contain. Our framework is based on learning identity correlations between profile attributes in an OSN community and on collecting ratings from OSN community members to evaluate the trustworthiness of target profiles. Our system guarantees utility, user anonymity, impartiality in rating, and operability within the dynamics and continuous evolution of OSNs. In this article, we detail the system design, and we prove its correctness against these claimed quality properties. Moreover, we test its effectiveness, feasibility, and efficiency through experimentation on real-world datasets from Facebook and Google+, in addition to using the Adults UCI dataset.

Scam Augmentation and Customization: Identifying Vulnerable Users and Arming Defenders

“…if a sender introduces him/herself as a recruiter, the signature should match it). For the company names, we used GlobalTrade website 4 and randomly chose 12 companies and their field of work to use in our grammar. A total of 100 emails were created and we randomly chose two of them for this study.…”

“…Researchers proposed several methods that can be used to steal information from users of social networks. Identity theft attack [4,9,34] and social engineering attack [35] are two types of these attacks. These attacks are variants of traditional security threats that leverage social networks as a new medium.…”

“…https://www.glassdoor.com/index.htm 2 https://www.indeed.com/ 3 http://www.419scam.org/419representative.htm4 https://www.globaltrade.net/global/Sales-and-Distribution/expert-serviceprovider.html…”

Knowledge‐based approaches for identity management in online social networks

“…This has been studied from different perspectives, with both automated and human‐assisted techniques, but the focus has generally been on formulating the sybil defense mechanism as a classification problem of OSN accounts as either honest or sybil (Kansara & Shekokar, ). More recently, another body of works has proposed a new formulation based on the evaluation and validation of OSN accounts, based on how consistent they look with respect to the majority of the profiles in the OSN (Bahri, Carminati, & Ferrari, , ; Soliman, Bahri, Carminati, Ferrari, & Girdzijauskas, ). To provide a high level view of this landscape, with also a summary of pointers to related works in the literature, we provide Figure as a reference starting point for comprehending the existing techniques.…”

“…This is basically due to the fact that people tend to connect and socialize with those who are similar to them, building as such a notion of community-level social identity (Bahri et al, 2014). As an extension to the work of Bahri et al (2014), Bahri et al (2016) presents a privacy preserving version of this proposed identity validation system based on correlated attributes. The privacy issue comes from the fact that raters (i.e., members of the community who evaluate the correlated attributes) have access to parts of the profiles that are being evaluated.…”