This article focuses on the credibility analysis of human resource attendance data, aiming at solving the credibility problem faced by enterprises in attendance data processing. In this article, a method combining image enhancement and feature extraction is proposed, and the corresponding algorithm model is constructed. Firstly, the original attendance data is preprocessed, then the image is enhanced by CLAHE (Adaptive Histogram Equalization), then the feature extraction is completed by combining PCA (Principal Component Analysis) and LBP (Local Binary Pattern), and finally the credibility analysis is carried out by using SVM (Support Vector Machine). The results show that during the training process, the accuracy of the model is finally stable at 95%, the recall rate is stable at 92%, the F1 value is 93.5%, and the MSE (Mean Square Error) is reduced to 0.03, and all indicators are better than the traditional simple statistical rule model. This achievement proves that the fusion method effectively improves the accuracy and reliability of the credibility analysis of human resource attendance data, and provides a more powerful decision support means for enterprise human resource management.
Zhu et al. (Sun,) studied this question.