Authors : Melam Divya Varalakshmi; Siram Sruthi; Ungarala Kusumanjali

Volume/Issue : Volume 9 - 2024, Issue 3 - March

Google Scholar : https://tinyurl.com/yjcccrxk

Scribd : https://tinyurl.com/3m2sercs

DOI : https://doi.org/10.38124/ijisrt/IJISRT24MAR263

Abstract : Fake profile detection is a critical task in maintaining the authenticity and safety of online communities. With the rising predominance of web- based entertainment stages, the presence of phony profiles has turned into a worry. This abstract explores are various techniques used to detect fake profiles including analysing user behaviour checking for inconsistencies in profile information, and employing machine learning algorithms. The detection process involves analysing patterns and anomalies in user data to identify suspicious activity. AI calculations assume a huge part in this cycle by gaining from marked datasets of veritable and counterfeit profiles. These calculations can investigate highlights, for example, profile pictures, posting conduct, network associations, and commitment examples to make expectations on the realness of a profile. However, it's important to acknowledge that no detection method is foolproof. Fake profile creators constantly evolve their strategies to evade detection. Therefore, continuous monitoring, user feedback, and updates to detection algorithms are necessary to stay ahead of these malicious actors. The ongoing efforts to detect and combat fake profiles contribute to creating a safer and more trustworthy online environment. By leveraging various techniques and advancements in machine learning, platforms strive to maintain the integrity of their user base and protect their users from potential scams and fraudulent activities.One example of a machine learning algorithm used for fake profile detection is the Random Forest algorithm. Irregular Backwoods is a well known troupe learning technique that joins various choice trees to make expectations. With regards to counterfeit profile recognition, the Irregular Backwoods calculation can be prepared on a dataset that incorporates both certified and counterfeit profiles. The calculation gains from different elements like profile data, posting conduct, network associations, and commitment designs. I trust this theoretical furnishes you with a succinct outline of phony profile location!

Keywords : Web-based Entertainment , Counterfeit Profiles , Irregular Woodland.