The invention relates to the field of artificial intelligence, and provides a conversation mode-based verbal skill recommendation method and device, equipment and a medium. Sentiment classification is automatically performed based on a sentiment classification model obtained by training without manual intervention, and verbal skill recommendation and dialogue mode recommendation can be automatically performed based on a constructed verbal skill-dialogue mode recommendation model. Verbal skill recommendation and dialogue mode recommendation are converted into click rate estimation problems to assist a customer service in better communication with a customer, emotion of the user is analyzed in real time according to real-time dialogue contents of the user and the customer service, and verbal skill and dialogue modes are recommended in real time according to the emotion of the user and answers of the customer service, the traditional question and answer problem is solved in a search recommendation mode, the answer quality of customer service staff is effectively improved, more satisfactory services are provided for users, customer stickiness is increased, and the customer complaint rate is reduced. In addition, the invention also relates to a block chain technology, and the trained model can be stored in a block chain node.