Research On Behavior Recognition Of Hone-Based Elderly Care Based On Wearable Sensor Data

In view of the current situation of home-based elderly care services with complex content,high work intensity,extensive management methods,imperfect evaluation systems,and low management efficiency,this paper discusses how to carry out data analysis and information management of nursing service work from a data-driven perspective.In this paper,based on the sensor data of wearable devices,the research on behavior recognition of home care for the elderly is carried out,which provides decision-making basis for the analysis of nursing service work content and job performance evaluation.First of all,by investigating the content of home care services for the elderly,it analyzes the behavior and execution characteristics of nursing services.Secondly,the wearable device wearing and sensor data collection scheme is designed according to the characteristics of nursing behavior.And a data-driven nursing behavior recognition research plan is designed,including preprocessing of nursing behavior data collected by sensors,time series data segmentation,feature extraction using time domain,frequency domain and other relevant physical features.Finally,using the collected volunteer data,the nursing behavior recognition scheme based on wearable sensor data is analyzed and discussed.The experimental results show that the recognition accuracy,recall rate and f1-score of 9 nursing activities are better,all above 90%,and the highest can reach 1.0.Through the results,we can find:(1)The standardized data is more valuable for the research of nursing behavior identification.(2)Different nurses have received unified training and will have inconsistent understanding of action points,which will lead to differences in recognition accuracy.(3)Mixing the normative data of the trained personnel and the rough data of the untrained personnel for training can obtain a model with better generalization ability and recognition accuracy.To sum up,wearable devices have high recognition accuracy in nursing behavior recognition research,and standardized data is more meaningful for nursing behavior recognition,which can significantly improve the management efficiency of nursing staff and further improve the quality of nursing staff’s service.This study actively explores the use of information technology and artificial intelligence theory to solve practical application problems,which can effectively promote the informatization and intelligent development of the elderly care service industry,and provide exploratory research references for the rapid development of the elderly care service industry.