Recommendations in Social Media Applications to Ensure Personification and Safety using Machine Learning

Abstract

Myriads of social media utilization lead to various issues like personalization hacks, data security problems, and safety. A recommendation is of paramount importance to alleviate this problem when there is a huge amount of data and the number of participants on the platform is increasing exponentially. Unfortunately, modern social media research has enhanced the performance and personalization of recommendations in many fields, yet largely underutilizes the power of artificial intelligence to enable personalized recommendations system for social media platforms like WhatsApp, Facebook, Twitter, etc. With advancement inside the global of technology every hour and every day new features are delivered to the list. In a manner, social platforms are merging into our actual existence, and to achieve personification and related safety, users can get any one safety factor from all 6 classes with this approach. This factor provides the basis for personification and the implementation of safety precautions.
 This research proposes recommendations for personification in social media applications. The proposed Modified Inception Resnet V4 Convolutional Neural Network (MInReCNN) outperforms embedded media persona analysis and classification through text, image, and video data. Using these prediction classes better decisions can be made in given social media domain.