Tracking Challenges in Online Social Environment using Deep Learning Techniques

Abstract

Social network event prediction is much more important task in many of the applications like medical, security, etc. With fast-growing technology, People spent most of the time in Social Networks. They can express their views and opinions in social network community. The main reason behind this phenomenon happens to be the ability of online community. It can provide a platform for users to connect with their family, friends, and colleagues. The information shared in social network and media spreads very fast, which makes it attractive for attackers to gain information. However, event prediction is a more complex task because it is challenging to classify, contains temporally changing the concept of discussion and heavy topic drifts learning. In this research, we present to addresses the challenge of accurately representing relational features is observed from complex social communication network behavior for the event prediction task. In this, graph learning methodologies are more complex to implement. Here the concept gives, to learn the complex statistical patterns of relational state transitions between actors preceding an event and then, to evaluate these profile findings temporally. The event prediction model which leverages on the RFT framework discovers, identifies and adaptively ranks relational occurrence as most likelihood predictions of event in social network communities. Most extensive experiments on large-scale social datasets across important indicator tests for validation. It shows that the RFT framework performs comparably better by Hybrid Probabilistic Markovian (HPM) predictive method. Deep learning relational models appear to have considerable potential, especially in the fast growing area of social network communities. This study opens the door to precise prediction events in spatio-temporal phenomena, adding a new tool to the data science revolution. Also, Social network analysis software has many algorithms for graph features data has been collected.