At present, the variation coefficient is used to calculate the difference between personnel and individual posts, which leads to the poor ability of allocating enterprise human resources. Therefore, a high-quality enterprise human resource allocation method based on intelligent big data is proposed. According to the four principles, the process of model building is designed, the hypothesis of model building is made, and the model of optimal allocation of human resources is constructed. From the personnel and post quantity, personnel and post structure, personnel and post quality, system, and other four aspects of optimal allocation, design the model running process, network personnel and posts, and use network indicators to match personnel and posts. The adjustment layer is introduced into the basic CNN model structure, and the input layer of intelligent big data is used to normalize the score of current high-quality human resources of enterprises. After the local characteristic processing is implemented on the human resource scoring matrix, the human resource allocation results are output to realize the allocation of high-quality human resources of enterprises. The experimental results show that the evaluation effect of the method is good, the evaluation time is short, the work efficiency is high, the credibility of human resource structure allocation is good, and the human resource allocation method has a good human resource structure allocation ability.