A business processing method, apparatus, and computer-readable storage medium

Abstract

The present application provides a service processing method, apparatus, and computer-readable storage medium. The method includes: a service processing apparatus obtains a service to be processed, where the service includes a user identifier and a description text of the service; based on a pre-trained solution diagnosis model, determines a solution corresponding to the service, and the solution diagnosis model is trained based on historical services and their solutions; processes the service based on the above solution. Therefore, before the service is assigned to a professional technician, a solution can be provided through the solution diagnosis model, which can reduce the amount of services assigned to professional technicians, is beneficial to improving the efficiency of service processing, and thus enhances the user experience and satisfaction.

Claims

1. A business processing method, characterized in that, it includes: Obtain the business to be processed, where the business includes a user identifier and a description text of the business; Determine the similarity between the description text of the business and the description texts of multiple historical businesses respectively; Determine that the similarity between the description text of the business and the description texts of the multiple historical businesses is less than a preset threshold; Return the description text of the business for modification until the similarity between the description text of the business and the description text of at least one historical business among the multiple historical businesses reaches the preset threshold; When the similarity between the description text of the business and the description text of at least one historical business among the multiple historical businesses reaches the preset threshold, input the description text of the business into a solution diagnosis model; Based on the pre-trained solution diagnosis model, determine the solution corresponding to the business, where the solution diagnosis model is trained based on historical businesses and their solutions; Process the business based on the solution.

2. The method according to claim 1, characterized in that, The processing the business based on the solution includes: Query in the database whether the user data corresponding to the user identifier conforms to the business scenario corresponding to the solution; When the user conforms to the business scenario, call the relevant interface of the operation and maintenance tool corresponding to the solution to process the business.

3. The method according to claim 1, characterized in that, The method further includes: When the processing of the business based on the solution fails, based on a pre-trained classification model, determine the business classification corresponding to the business, where the classification model is trained based on historical businesses and their corresponding business classifications; Based on the business classification, assign the business to the corresponding technical personnel.

4. The method according to claim 3, characterized in that, The solution diagnosis model and the classification model are respectively obtained by training a text convolutional neural network TextCNN model.

5. The method according to claim 1, characterized in that, The determining the similarity between the description text of the business and the description texts of multiple historical businesses respectively includes: Based on the term frequency TF - inverse document frequency IDF algorithm, generate a first word frequency vector corresponding to the description text of the business, where the first word frequency vector indicates the characteristic words in the description text of the business; Based on the first word frequency vector and the second word frequency vectors corresponding to the description texts of each historical business among the multiple historical businesses, calculate the similarity between the first word frequency vector and each second word frequency vector.

6. A business processing device, characterized in that, it includes: A business acquisition module, configured to acquire the business to be processed, where the business includes a user identifier and a description text of the business; A similarity matching module, configured to determine the similarity between the description text of the service and the description texts of multiple historical services respectively; determine that the similarity between the description text of the service and the description texts of the multiple historical services is less than a preset threshold; return the description text of the service for modification until the similarity between the description text of the service and the description text of at least one historical service among the multiple historical services reaches the preset threshold; in the case where the similarity between the description text of the service and the description text of at least one historical service among the multiple historical services reaches the preset threshold, input the description text of the service into a solution diagnosis model; A solution diagnosis module, configured to determine the solution corresponding to the service based on a pre-trained solution diagnosis model, where the solution diagnosis model is trained based on historical services and their solutions; A solution processing module, configured to process the service based on the solution.

7. A service processing device, characterized in that, it includes a processor and a memory, the processor is coupled with the memory, the memory is used to store a computer program, and the processor is used to execute the computer program so that the device implements the method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, it includes a computer program, and when the computer program runs on a computer, it causes the computer to execute the method according to any one of claims 1 to 5.

Images