One of the primary worries for the loved ones of elderly individuals is the accidental fall that can sometimes even lead them to death. Recent studies have shown that sensor-based methods in identifying falls mark a significant advancement in proactive fall detection, facilitating prompt assistance from concerned family members or caregivers to aid elderly individuals. Our primary objective revolves around the design and refinement of an arm-worn sensor-based fall detection method. Intelligently integrating accelerometer and gyroscope data into a wearable device, the model can efficiently analyze variations in different human daily activity movement patterns. This multifaceted approach enables our model to discern with precision between falls and activities of daily life (ADL). Through extensive experimentation and analysis, our results demonstrate the effectiveness of the proposed model in accurately identifying fall events while minimizing errors in detecting ADL. Achieving an impressive accuracy score of 98.66%, our novel approach proves to be more effective and beneficial in the realm of fall detection among elderly individuals compared to existing state-of-the-art methodologies.