VOL. 13, NO. 24, DECEMBER 2018 ISSN 1819-6608 ARPN Journal of Engineering and Applied Sciences ©2006-2018 Asian Research Publishing Network (ARPN). All rights reserved. www.arpnjournals.com 9375 CROSS-PLATFORM RECOGNISATION OF UNKNOWN IDENTICAL USERS IN MULTIPLE SOCIAL MEDIA NETWORKS N. Naga Priyanka, N. Geetha and A. Viji Amutha Mary Department of Computer Science and Engineering School of Computing Sathyabama University, Chennai, India Email: vijiamumar@gmail.com ABSTRACT From very recent past years have witnessed the requirement and evolution of a vibrant research Crew on a large variation of online Social Media Network (SMN) platforms. Recognizing anonymous, same yet identical users among multiple SMNs is still a major problem. Clearly, saying that cross-platform exploration may help solve too many problems in social computing in both theory and applications. Up to now public profiles can be duplicated and easily impersonated by users with different purposes, most current user identification resolutions, which mainly focus on text mining of users ‘public profiles, fragile. Some studies have attempted to match users based on the location and timing of user content as well as writing style. However, the locations are sparse in the majority of SMNs, and writing style is difficult to discern from the short sentences of leading SMNs such as, Sina Micro blog and Twitter. Moreover, up to now online SMNs are quite symmetric, existing user identification schemes based on network structure are not effective. The real-world friend cycle is highly individual and virtually no two users share a congruent friend cycle. So that, it is more accurate to use a friendship structure to analyze cross-platform SMNs. Up to now anonymous users were influenced to set up similar friendship structures in the different SMNs, here they proposed the Friend Relationship-Based User Identification (FRUI) algorithm. FRUI Algorithm calculates a match degree for all candidate User Matched Pairs (UMPs) only, UMP with top ranks are considered as identical users. We also developed two propositions to improve the efficiency of the algorithm. The Results of these extensive experiments demonstrate that FRUI performs much better than current network structure-based algorithms. Keywords: friend relationship algorithm, user identification, cross-platform, social media network, anonymous identical users. 1. INTRODUCTION In the past decade, there are many types of social networking websites have started and contributed immensely to too many volumes of real-world data on people’s behaviors. Twitter was the 1, the largest micro- blog service, has more than 650 million of users and produces more than 375 million tweets per day [1]. Sina Micro-blog 2, the primary Twitter-style Chinese micro- blog site, has more than 555 millions of accounts and generates over 100 million tweets per single day [2]. Due to that diversity of various online social media networking sites (SMNs), people interested to use different SMNs for different purposes. For instance, RenRen 3, a Face book- style but antonymous SMN, is used in China for blogs, while Sina Micro-blog is used to share statuses. In other means, for every existent SMN fulfils some user requirements. In other terms of SMN management, matching these anonymous users across different SMN platforms that can be provide integrated components on the each user and inform corresponding regulations, such as targeting services of provisions. In the theory, these cross-platform explorations will allow a bird’s-eye view of all SMN user behaviours. However, those nearly all recent SMN-based studies that focuses on the single SMN platform, yielding incomplete data. User identification is also known for user recognition; user identity resolution is a, user matching, and the anchor linking program. Although there is no solution that can recognise all the identical anonymous SMN users, where some SMN elements may be used to recognise a portion of users across multiple SMNs. Many studies up to now have addressed the user reorganisation problem by examining public user profiles attributes, including screen name, birth-day, place, gender, profile photo, etc. [3], [4], [6], [7], [9], [10], [11], [13], [14], [15], [16], [17]. Up to now these attributes do not require exclusivity and are easily copied by users to different purposes (including malicious users), these schemes were regularly fragile. Some researchers have to leverage public user activities are to recognize users using post time, location & writing style [18], [19], [20], and [21]. Up to now location data is difficult to obtain and writing style is difficult to extract from short these sentences, these techniques were plagued by few limitations. Even though connections can be collected and are difficult to impersonate in nearly all SMNs, our literature review revealed only a few studies that explored in employing many user friends to recognise their users [22], [23], [24]. For the terms of data security and privacy, Narayanan [22] de-anonym zed a social network graph by correlating it with known identities. NS was the first effort to recognize users purely by using connections, and successfully matched 30% of the accounts with a 12% error rate. [23] Proposed a new Joint Link-attribute Algorithm (JLA) for match two social networks and obtained a new portion of identical users. The all new SMN connections fall into two categories: single- following connections as well as mutual-following connections. Single-following connections are also called following relation-ships or following links. If user A follows user B, then user A and user B have a following relationship (single-way fans in which one knows the other, but not vice versa). Following relationships are common in micro-blogging SMNs, such as Twitter and