In order to conduct an offence, malevolent attacks like spamming, phishing, or hacking are considered into cyber crime. The computer systems are deftly hacked and compromised, causing significant financial loss that might huge impact. Email spamming is the most well-known type of cyber attack since it uses up more cyber resources, such as memory, computing power, network bandwidth, traffic abuse, etc. Spam emails are mass-produced, unsolicited commercial emails that are sent for a variety of reasons. Studies show that more than 85% of today's email is spam. Researchers have come up with a number of strategies to control email spam, but some of them have been comprehensive or successful. The methods for detecting email spam that are currently in use have some major drawbacks. First of all, it has not been possible to efficiently separate spam emails from legitimate ones. This has increased the amount of false positives and false negatives, which has reduced the accuracy of detection. Second, when the volume of emails received rises, it takes longer to identify spam emails. Thirdly, because the detection filters are installed on the server, the server is overworked while handling large operations. Therefore, effective reaction mechanisms and efficient collaborative detection techniques are found for early, widespread identification and mitigation of spam emails and their source at the receiver side. User Authorization Phase, Feature Extraction Phase, Classification Phase, and Similarity Detection Phase are the four steps of an unique Probabilistic EShield Protocol (PEP) that provides Email Spam Detection with extra features. By evaluating the email content utilizing extra functions, PEP filters the spam email as well as the illegal sender of the incoming email. The results of experiments conducted for PEP demonstrate that PEP outperforms CRVSM in terms of detection accuracy, false positive and false negative rate reduction, and detection rate attainment. Therefore, three unique protocols have been presented in this study for email spam detection, offering cyber security in the cyber- space: Collaborative Reputation-Based Vector Space Model (CRVSM), Probabilistic EShield Protocol (PEP), and Optimized Feature Selection Protocol (OFSP).