For sharing of important and official information, email is used as a default medium of communication. Most of the institutions and companies prefer to use emails over all other mediums as it is one of the cheapest, easy to use, easily accessible, most official and reliable way of sharing information. It is used widely as it also provides the confidentiality of the data shared. But with the pros also comes the cons, as many people misuse this reliable and easy. way of communication by sending unwanted and useless bulk messages for their own personal benefits. These unwanted emails affect the normal user to face the problems like flooding of the mail box with unwanted emails making it harder to look for the useful once, even sometimes one may skip through important and useful emails because of all these unwanted emails. So, this gives rise to a need of a strong email spam detector which can filter maximum amount of spam emails with a greater accuracy so that a genuine email does not get filtered as spam. In this paper an integrated approach using Naïve Bayes algorithm along with Particle Swarm Optimization is used for email spam detection. Naïve Bayes algorithm is used for learning and classification of email as spam and ham. Particle Swarm Optimization is a stochastic optimization technique and is used for heuristic global optimization of parameters of Naïve Bayes. For experimentation Ling Spam dataset is considered and result is evaluated in terms of precision, f-measure, accuracy and recall.