A K-Means Clustering Algorithm for Data Analysis of Wearable Equipment of Construction Personnel

Wearable device data analysis is the key to construction personnel management. In the process of wearable device data analysis, signal interruption is easy to occur, which reduces the accuracy of analysis, and causes key data loss and abnormal signal interference. This paper proposes a k-means clustering method to analyze the similarity of wearable device data, enhance the transmitted signals and shorten the interval between adjacent signals. Then, the signal attenuation and frequency are analyzed iteratively. Finally, the wearable device signals are searched by clustering theory, and the final analysis results are output. MATLAB shows that k-means clustering method can analyze data accurately, the data loss rate is less than 10%, and the analysis accuracy rate is more than 80%, which is better than single clustering analysis method. Moreover, in the k-means clustering method, the data analysis time of wearable devices is shorter, the location analysis time is 1/2, 1/4, and 1/6, and the abnormal signal recognition time in individual devices, group devices, and all devices are better than the single clustering method. Therefore, k-means clustering method can meet the data analysis requirements, and is suitable for construction personnel management.