Model training method and device, computer equipment and readable storage medium

Abstract

The invention provides a model training method and device, computer equipment and a readable storage medium, and the method is used for an intention recognition model and comprises the steps: obtaining a target intention; wherein the target intention is obtained by performing intention recognition on a dialogue; querying a set reply matching the target intention according to the target intention;determining semantic similarity between the matched set reply and the following text of the dialogue; wherein the conversation following text is continued after the conversation preceding text; and ifthe semantic similarity is greater than a set similarity threshold, labeling the dialogue preceding text by adopting the target intention to obtain a first positive sample, so that an intention recognition model performs intention training according to the first positive sample. The intention in the first positive sample can be automatically recognized and labeled, and the labeled first positivesample is adopted to automatically train the intention recognition model, so that manual intervention is not needed, the optimization period of the intention recognition model is shortened, and the labor cost is saved.

Description

technical field [0001] The present application relates to the technical field of artificial intelligence, and in particular to a model training method, device, computer equipment and readable storage medium. Background technique [0002] In the process of telemarketing, the intelligent telemarketing robot can identify the real intention of the customer, and recommend a recommended reply that matches the real intention of the customer to the salesperson, so that the salesperson can refer to the above-mentioned recommended reply to communicate with the customer. However, the premise of the above method is that the intent recognition model needs to be built in the smart telemarketing robot. After the intent recognition model is put into operation, in order to improve the accuracy of the recognition results of the intent recognition model, the intent recognition model needs to be optimized. [0003] In the prior art, after the sample data is obtained, the real intention of the c...

AI Technical Summary

Problems solved by technology

[0005] This application proposes a model training method, device, computer equipment, and readable storage medium to automatically identify and label the intent in the first positive sample, and use the labeled first positive sample to automatically identify the intent Model training can shorten the optimization cycle of the intent recognition model without manual intervention, and save labor costs. It is used to solve the time-consuming and labor-intensive technology of the prior art through manual labeling, and the optimization cycle of the intent recognition model is long. question

Images