Revolutionizing Big Data Assessment for Human Activity Recognition with Metapath Context and Bi-directional Cascade Networks

Abstract

Human activity recognition involves automatically identifying and categorizing human activities using sensor data or video inputs. Challenges include interpreting complex activities, handling imbalanced data classes, ensuring privacy in video-based methods, and achieving efficiency with limited computational resources. This research, the Metapath Contexed Convoluted Bi-directional Cascade Network (MCCBCN) model is introduced, which leveraging metapaths to enhance accuracy and robustness in capturing contextual information for human activity recognition. By leveraging both convolutional and bi-directional cascade architectures, MCCBCN significantly advances the field allowing it to capture temporal dependencies effectively. The integration of the Enhanced Football Team Training Optimization Algorithm (EFTTOA) improves training efficiency and enhances the model's ability to capture complex temporal and contextual dependencies in human activity data. Additionally, the utilization of Large-scale Synthetic Minority Over-Sampling Technique (LSMOST) for dataset expansion ensures a more comprehensive and balanced representation of minority classes, mitigating biases from imbalanced class distributions and improving model generalizability. The proposed method achieved impressive evaluation metrics with an accuracy of 99.50%, F1-score of 99.9%, precision of 99.42% and recall of 99.3%, outperforming existing techniques. Also, the proposed MCCBCN time complexity as 10.2 seconds.