Abstract:In order to solve the problems of low accuracy and long processing time in the document distribution of nuclear power plant, this paper proposes a set of automatic and intelligent document distribution system, and researches the intelligent document distribution based on rules and machine learning. Through the identification and extraction of key information, the rule matrix based on key information is established, and the rule execution feedback mechanism is added to improve the rule matrix. In addition, the data source and data storage of the system are analyzed and designed, the distribution rules are improved by machine learning, and the document recommendation list for the system is calculated by using the algorithm. By adding rule engine and intelligent recommendation to the original distribution process, the automation and intelligence of the system are greatly enhanced. From the practical application, the intelligent document distribution system can effectively improve the accuracy and timeliness, and realize the leap from artificial to intelligent. In order to further improve the application effect, this paper proposes the following improvement direction from the perspective of language algorithm model and deep learning framework.