Abstract:- Phishing attack is a simplest way toobtain sensitive information from innocent users. Aim of the phishers is toacquire critical information like username, password and bank account details.Cyber security persons are now looking for trustworthy and steady detectiontechniques for phishing websites detection. This paper deals with machinelearning technology for detection of phishing URLs by extracting and analyzingvarious features of legitimate and phishing URLs. Decision Tree, random forestand Support vector machine algorithms are used to detect phishing websites. Aimof the paper is to detect phishing URLs as well as narrow down to best machinelearning algorithm by comparing accuracy rate, false positive and falsenegative rate of each algorithm. Keywords: Phishing, Feature Classification,Random Forest classifier, etc