Model self-iterative intelligent customer service quality inspection system and method with configurable business logic

Abstract

The invention discloses a model self-iterative intelligent customer service quality inspection system and method with configurable business logic. The model is composed of seven parts including the training module. Through model and rule setting, the entire service process is automatically detected; structured and unstructured content is analyzed in real time, and the model outputs quality inspection results in real time; the system can preset various A service early warning model, which reminds the service provider and the manager during the service process; the system is equipped with visual data display and analysis, so as to keep abreast of the distribution of quality inspection problems and the development trend of problems.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence, in particular to a model self-iterative intelligent customer service quality inspection system and method with configurable business logic. Background technique [0002] Customer service (hereinafter referred to as "customer service") is very important to the operation and development of an enterprise. Customer service quality inspection can help companies detect service quality, find service problems, and then better train skills and improve the service level of customer service personnel. With the development and expansion of enterprises, traditional manual quality inspection cannot cope with the ever-increasing business data, and the disadvantages of traditional quality inspection methods are becoming more and more obvious. different. [0003] Generally speaking, the traditional manual customer service quality inspection system will have the following four pain points: first, ...

AI Technical Summary

Problems solved by technology

[0010] 1. The data connection between the quality inspection system and the customer service system is highly complex;

[0011] 2. Single-point quality inspection cannot solve complex logic business;

[0012] 3. The accuracy of the quality inspection algorithm is relatively fixed, and once delivered, it cannot be further optimized;

[0013] 4. The quality inspection task is fixed, unable to adapt to new scenarios and new requirements, such as reconfiguring the quality inspection model for different business needs, adding new business scenarios or new quality inspection tasks, etc.;

[0014] 5. Deep learning model training generally takes a long time, and the GPU uses a lot of memory

Images