The thesis presented by PhD student Mohammed Khalaf Raheem, under the supervision of Professor Dr. Ali Hilal. This intriguing research explores the application of Deep Learning techniques to predict human activities using sensor data from four different public datasets. The researchers introduced an innovative model that pioneers the extraction and classification of features in this field. This novel approach incorporates an extraordinary method for digital signal processing, setting it apart from previous techniques. The study’s findings are truly remarkable! The implemented approach outperformed existing methods, showcasing superior results in the domain of human activity prediction. Moreover, the potential for real-world applications in health, human-computer interaction, and other related fields has been unveiled. Finally this thesis has been accepted successfully.