The training and testing dataset originates from non-invasive sensors, such as motion, magnetic, and temperature sensors, installed in households. This study proposes a non-invasive monitoring system to enhance physical safety and facilitate the early detection of behavioural changes, especially in individuals with mild cognitive impairment(MCI). Its objective is to identify anomalies in daily activities through convolutional autoencoder models, capable of recognising deviations from typical behavioural patterns. This dataset collection was funded by the European Union’s Horizon 2020 research and innovation programme under Grant Agreement No. 857159, project Smart & Healthy Ageing through People Engaging in Supportive Systems (SHAPES). It was also funded by MCIN/AEI/10.13039/501100011033 under Grant TALENT-BELIEF (PID2020-116417RB-C44) and the Project MIRATAR TED2021-132149B-C41 funded by CIN/AEI/10.13039/501100011033 and by the European Union NextGenerationEU/PRTR