A Comprehend Survey On Future Directions Of Artificial Intelligence Techniques For Social Behaviour Analytics

Abstract

As social media platforms and digital communication channels have grown at an unprecedented pace creating vast amounts of behavioural data, Social Behaviour Analytics has become a key area of Artificial Intelligence (AI) importance. This research looks at a number of AI approaches to human behaviour analysis and prediction such as Machine Learning (ML), Deep Learning (DL), Graph Neural Networks (GNNs), Explainable AI (XAI), and Reinforcement Learning (RL). The introduction reveals how these techniques can be applied to gain insight into user interaction, sentiment and social influence in diverse domains like the business world, healthcare, and social media platforms. The survey provides a general survey of the techniques developed over the past 6 years (2020-2025), in particular highlighting the adoption of state-of-the-art deep learning and graph-based models over classical ML models. Most existing research has been done on structured data classification and prediction, and recently, models using GNNs have been developed to model relation and multimodal learning to incorporate multiple data sources. There are advantages in terms of accuracy and performance, but problems remain such as data sparseness, computational complexity, and lack of interpretability and privacy issues. The research study is a step towards state of the art and offers insights into the needs of future social behaviour analytics systems with respect to scalability, explainability and adaptability.