Safe web browsing and feeding confidential information into websites require the use of protected and secured websites. For the web security, a number of anti-phishing tools have been proposed which provide web user with a dynamic system of warning and protection against potential phishing attacks. Earlier study shows that there is no anti-phishing tool gives satisfactory result in identifying the phishing web pages. For the solution of this problem, in this paper a Bayesian classification approach is proposed to identify the phishing websites. Bayesian filter require two datasets in their approach; one is legitimate website details and second thing is phishing website parameters. A large set of legitimate transactional websites are needed in the study because the set of websites mostly resembles just like phishing websites and the filter must have numerous examples of legitimate transactional websites to achieve a low false positive rate. With the use of Bayesian Classification, some prominent results obtained by selecting phishing indicators.