Author : Smita Sitaram Wagh and P. P. Joshi.

Today, the enormous growth and volume of online social networks and their features, together with the large number of socially connected users, it has become difficult to explain the true semantic value of published content for the detection of user behavior and anxiety. We propose a platform for analyzing the content of social media that finds anxiety for analyzing and detecting abnormal behavior that deviates significantly from the norm in large-scale social networks. Different types of analysis have been perform for a better understanding of various user behaviors in detecting highly adaptive anxiety users. We propose a new approach to the data extraction and classification process to contextualize large-scale networks appropriately as well as collect a significant number of user profiles from the activities of Facebook and Instagram. Comprehensive assessments were made of real-world data sets of user activity for both social networks.

Keywords: Sentiment analysis, Social Interaction, Recommendation system, social Media, Machine learning.