Bharat S. Borkar et al., International Journal of Advanced Trends in Computer Science and Engineering, 9(4), July – August 2020, 6214 – 6220 6214 ABSTRACT In the current scenario, online social media platforms are the technology's most common and fastest tools for information exchange. The majority of people from all backgrounds expand their time on social networking platforms. An enormous amount of information is developed and shared worldwide through social networks. Such motives have contributed to unauthorized participants engaged in malicious acts against members of the social platform. False account formation is seen on social media as doing more damage than in any other form of Cybercrime. This offense must be identified well before the consumer is told about both the development of the fake identity. Numerous algorithms and approaches have been suggested for the identification of false identities, most of which use the vast amounts of raw data produced by social platforms. In this research, we proposed fake identity detection of social accounts on twitter dataset. Various machine learning algorithms have been used to evaluate the proposed results using NLP techniques. SVM, Fuzzy Random Forest, and Naïve Bayes have used for classification. The experimental analysis shows the effectiveness of the system and how it produces better accuracy than other machine learning algorithms as well as existing systems. Key words : Machine Learning, Naïve Bayes, Fuzzy Logic, Random Forest, twitter dataset, fake identity, bots, Natural Language processing, classification. 1.INTRODUCTION The social media platforms are now gradually the domain of our lives and include different facets of the day-to-day societal activity. The demands of people perform certain social positions in such these as media prefer what they do in actual life. Behaviors, perceptions, acts, and habits in these systems, individuals hold (Social culture in Social Media) When the value of such media rises, this concept is more important to research and to test Comprise. The paper pays intellectual attention to the issue by focusing on regional cultural aspects Google app discrepancies, among the most common social media. Social networking is a level higher quickly these days, which is vital for ad strategies and celebrities that try to boost it by increasing their number of followers and fans. Nevertheless, false accounts, produced apparently on behalf of organizations or individuals, can destroy their credibility and decreasing their numbers of friends and comments. They are still suffering from bogus notifications and excessive ambiguity with others. Fake profiles of all kinds generate negative effects that counteract the potential of social media in marketing and promotions for companies and lay the groundwork for online harassment. In an online world, consumers have specific questions concerning their privacy. There are a few social media sites which include Twitter, Google+, Youtube, Instagram, Flickr, Facebook, and snap- chat. There were 823 million individuals who used social media on their smartphones every day that is an improvement over the previous fiscal quarter of 654 million these consumers. Social networking sites such as Facebook cannot yet offer real-time updates to false accounts, so for semi-technically advanced consumers it is impossible to differentiate between true so false accounts, In fact, other big data problems, namely data collection, how to manage data streams, and how to deliver instantaneous user replies, have to be addressed when running on vast quantities of data concurrently to obtain reliable profile recognition performance. Earlier work on fake accounts tackles experimentation that evaluates preventive measures against fake user behavior patterns. A Facebook social manipulation research project using the google maps API analysis quantifiable information about the number of men and women friends, data access documents, clustering algorithms of mutual acquaintances, details about education and work, location Facebook users knowledge, and mutual interests. The security measures to defend users from attackers include knowledge of data, privacy laws, techniques to enhance safety, and awareness-raising training. Identification of Fake Identities on Social Media using various Machine Learning Algorithm Bharat S. Borkar 1 , Dr. Manish Sharma 2 1 Department of Computer Science & Engineering, Gyan Vihar School of Engineering & Technology, Suresh Gyan Vihar University, Jagatpura, Jaipur, India, borkarbharat.sgvu@gmail.com 2 Department of Computer Science & Engineering, Gyan Vihar School of Engineering & Technology, Suresh Gyan Vihar University, Jagatpura, Jaipur, India, manish.sharma@mygyanvihar.com ISSN 2278-3091 Volume 9, No.4, July – August 2020 International Journal of Advanced Trends in Computer Science and Engineering Available Online at http://www.warse.org/IJATCSE/static/pdf/file/ijatcse299942020.pdf https://doi.org/10.30534/ijatcse/2020/299942020