Identification method and system for intelligent quality inspection of service forbidden words

Abstract

The invention relates to a natural language processing technology, in particular to an identification method and system for intelligent quality inspection of service forbidden words, and the method comprises: performing text transcription and text preprocessing, i.e., carrying out the letter capital and lower-case conversion and part-of-speech tagging of text data of customer service; performing keyword retrieval; performing text transcription error identification, i.e., judging whether text transcription is reasonable or wrong according to historical customer service session text data; performing text semantic analysis, i.e., analyzing semanteme of the text in the customer service session content hitting the service forbidden word keyword to obtain an identification result of whether thesupport staff uses forbidden words or not ; and filtering an identification result that the support staff does not use forbidden words to obtain a final inspection result. The context of the keywordsin the text and historical session text data are combined in the text analysis process, and the accuracy of quality inspection is improved through rule matching.

Description

technical field [0001] The present invention relates to the technical field of natural language processing, in particular to a method and system for identifying forbidden words in intelligent quality inspection services. Background technique [0002] Service quality inspection is a very important quality control link in the hotline service operation process. Traditional service quality inspection mainly relies on manual random inspection of daily agent recordings for follow-up and analysis. Formulate improvement plans and implement improvement plans for various problems found, so as to further improve the internal service quality control system, and objectively and truly reflect the rapid response and close cooperation between the front desk and the back office of the call center from the data. Look for service shortcomings, put forward suggestions and suggestions for improvement, and assist in business process improvement, so as to continuously improve the service level of...

AI Technical Summary

Problems solved by technology

At present, the quality inspection method of service ban in intelligent quality inspection is relatively single. It is often to directly search the service ban keyword for the content of the customer service conversation text. If the service ban keyword is retrieved, the customer service service is not standardized. In the absence of error and semantic understanding of the text, the quality inspection results are not accurate enough to meet actual needs

Images