Inspiration: Our team has seen many people including those in the tech industry being scammed by fake accounts online. As a group of students, we, Vansh Aggarwal, Tannmay Kumarr Baid, Ojas Ashish Kulkarni & Aditya Kathait, are deeply concerned about the escalating problem of fake social media accounts, we have united our passion and expertise to tackle this pervasive issue head-on. We recognise the insidious nature of fake profiles, often harbouring malicious intent through phishing links and catfishing schemes, and have witnessed the devastating consequences of cybercrimes perpetrated through these deceptive accounts.
What it does:
Our approach is grounded in a thorough understanding of the behavioral characteristics and patterns associated with fake social media accounts. We meticulously analyze the content, interactions, and engagement metrics of these accounts, identifying red flags and anomalies that deviate from the norms of genuine user behavior. Harnessing the power of machine learning and artificial intelligence, we are developing sophisticated algorithms capable of deciphering the subtle cues that betray the true nature of fake accounts. These algorithms will learn and adapt, continuously refining their detection capabilities to stay ahead of the ever-evolving tactics employed by malicious actors.
How we built it:
Our project mainly consists of a website we made using the express framework of nodejs. This acts as a domain where the user can access the main model. The model has been built on tensorflow using a neural network model. The reason we used neural network is because it provides a 2d aspect which is often needed to predict on human based data. Further more we also have a google extension that one can use to access this model.
Challenges we ran into:
The challenges we faced were integration of model and the websites. Some of these challenges have been solved while some are still in dev phase. Furthermore, predicting on human data often gives less accuracy.
Accomplishments that we're proud of
We are proud to say that we have enabled a browser plugin that can detect fake onine accounts which is a massive win for our team as it enables a user to individually detect fake accounts. We are also proud of our algorithm which uses AI and ML
Log in or sign up for Devpost to join the conversation.