Vaibhav V. Satane and Rohit N. Devikar/ Elixir Inform. Tech. 86 (2015) 35183-35186 35183 Introduction Phishing is type of online attack whose target is influencing the online user as well as found the weakness in system processes as caused by online users. Worldwide Phishing is one of the cyber security threats. Phishing is the act of convincing users to give up their personal information, either. Phishing is an action of stealing user’s financial details. In website phishing the fraud user creates website which is similar to original famous websites such as Gmail, Twitter, Dropbox, Facebook and Paypal. The fraud user creates URL with including the name of these famous websites and URL send to the user through social networking sites due to the use of social networking site increased nowdays so social networking sites are most popular to phisher. According to PhishTank, the definition of phishing attack is that “Phishing is a wrong effort, commonly made with the help of email, to pinch user personal information. According to Colin Whittaker, phishing is accessing website without permission and use behalf of a third party with the purpose of baffling viewers into performing an action with which the viewer would only trust a true agent of the third party [10]. Data mining is extraction of useful patterns from data source like data warehouse, data repository and database. Patterns are effective, unique and reasonable. Data mining are also known as knowledge extraction, data dredging, knowledge discovery and data mining. Phishy URL and Social networking site Online Social network services such as YouTube, Facebook, Twitter, and MySpace, have recently popular due to supportive information platforms that allow users to share and to interact with other user. Social networking sites are one of the main ways for users to keep track and communicate with their friends online. The increase in popularity of social networking sites allows them to collect a huge amount of personal information about the users, their friends, and their lifestyles Twitter Social Network Twitter is a much simpler social network than Facebook and MySpace. It is designed in such a way that where users send short text messages (i.e., tweets) that appear on their friends’ pages. Unlike Facebook and MySpace, no personal information is shown on Twitter pages by default. Users are identified with the help of name of user. Only register users are able to read the tweets and post the tweets. A Twitter user can start “following” any other user such as famous and respectable people from different area like sport, politics, movie industry, businessman, different private or public organization people and their friends also. Authorized user are the register user who has twitter account. Authorized user receives tweets the other user’s on his page. In twitter social networking site tweets are visible to everyone by default. Other authorized user can retweet through twitter their website. Related work Chia-Mei Chen et. al proposed a suspicious URL identification system for use in social network environments is proposed based on Bayesian classification. The proposed system consist 3 modules the first module is data collection, posts are collected including time and content. Posts that lack URL information are considered benign. In the second module, feature extraction, the features are retrieved and a feature vector is constructed for classification. In the third module, the Bayesian classification model, posts are classified based on a pertained classification model [1]. Neda Abdelhamid et.al proposed the Multi-label Classifier based Associative Classification algorithm to detect phishing website also want to identify features that distinguish phishing websites from legitimate ones.it consist no of steps (1) the end- user clicks on a link within an email or browses the internet. (2) Then directed to a website which is genuine or phishy. (3) A script written in PHP that is embedded within the browser starts processing to extract the features of the test data and saves them in a data structure. (4) Now, the intelligent model will be active within the browser to guess the type of the website based on rules learnt from historical websites (previous data collected). The rules of the classifier are utilized to predict the type of the test data based on features. If the website is recognized as genuine no action will be taken. On the other hand, when the ABSTRACT Phishing is considered a form of web threats that is defined as the art of mimicking a website of an honest enterprise aiming to obtain user’s confidential credentials such as usernames, passwords and social security numbers. Social engineering and technical tricks are commonly combined together in order to start a phishing attack. The phishing attack starts by uploading the post on social network site which contain fraud URL that seems authentic to potential victims advising them to meet site and update or validate their information by following a URL link. The fraud user posts their comment on social network service that contains URL which are malicious. These malicious URL some time it contains malware or virus. When user clicks on these fraud URL Then viruses enter into user system and system gets affected. The proposed system Extraction of feature set for detecting fraud URL using ANN in social networking services effectively detect fraud URL with high accuracy. © 2015 Elixir All rights reserved. Elixir Inform. Tech. 86 (2015) 35183-35186 Information Technology Available online at www.elixirpublishers.com (Elixir International Journal) Phishing URL Detecting Using ANN Classification in Online Social Network Vaibhav V. Satane and Rohit N. Devikar Department of Information Technology, AVCOE Sangmner,District:-A.Nagar Maharashtra, India. ARTICLE INFO Article history: Received: 1 August 2015; Received in revised form: 12 September 2015; Accepted: 18 September 2015; Keywords Phishing, Data Mining, Classification Artificial Neural Network, Phishtank, Online Social Network. Tele: E-mail addresses: satanevaibhav@gmail.com © 2015 Elixir All rights reserved