Web calling services are exposed to numerous social security threats in which context of communication is manipulated. A attacker establishes a communication session to send numerous simultaneous pre-recorded advertisement calls (Robocalls), distribute malicious files or viruses and uses false identity to conduct phishing. User identification alone is not sufficient to provide a high level of trust between communicating participants. Therefore, we propose 'ProtectCall' a trust model that allows web calling services to estimate the trustworthiness and reputation of their users based on the evaluation of three parameters: authenticity, credibility and popularity. The main objective of ProtectCall is to protect web communication services from social security threats. ProtectCall allows users to make decisions based on the trustworthiness of their communicating participants