An Intelligent Sport and Fitness Monitoring Based on Improved Quantum-Inspired Differential Evaluation with Dendritic Neural Networks

In recent years, the intelligent sports and fitness monitoring influences data analytics for producing the real-time perceptions into the performance of athlete and fitness, however it faces difficulties includes privacy concerns, data accuracy and incorporating with existing systems. In this research, Improved Quantum-inspired Differential Evaluation with Dendritic Neural Network (IQDE-DNN) is proposed to increase convergence speed. Initially data is collected from the hardware sensors such as gyroscope, skin temperature, optical heart rate and others sensors. Further data is pre-processed by utilizing the Empirical Mode Decomposition (EMD) to decompose the nonlinear and stationary signals from the time series to residuals. Then the initial weights and threshold values are generated by using the Quantum Inspired Differential Evolution optimization (IQDE) to increase the performance of the evaluation. Further Dendritic Neural Networks (DNN)is employed to evaluate the data from the soma layer as fitness and sport of an athlete. From the results, the proposed IQDE-DNN established better results when compared with existing Internet of Things based Cyber Physical systems of Physical Fitness Evaluation Method (IoT-CPS-PFEM) in terms of accuracy (95.4%), precision (90.3%), recall (92.1%) and F1-Score (90.1%) respectively.