Recommendation System For Social Platform Using Deep Learning Techniques

This paper explores advanced techniques in developing a Friends Recommendation System on social media platforms, specifically Facebook. By leveraging user behavior data such as the number of followers, followings, and mutual friends, the system aims to provide accurate friend suggestions. This recommendation methodology can be applied across various domains, including search engine recommendations (Google, Explorer, Microsoft Edge), music recommendations (Wynk), video recommendations (YouTube), movie recommendations (Amazon Prime), and product recommendations (Flipkart, Amazon). The study employs similarity coefficient calculations like Jaccard Distance and Cosine Similarity, alongside ranking measures such as PageRank. Furthermore, the system's performance is evaluated using advanced deep learning techniques, including Graph Neural Networks (GNN), Convolutional Neural Networks (CNN), and Recurrent Neural Networks (RNN). Comprehensive experimentation involves measuring the F'l-Score and comparing the accuracy of various machine learning and deep learning algorithms to identify the most effective methods for friend recommendations.