NLP-based operator work order intelligent processing method and system

Abstract

The invention provides an NLP-based operator work order intelligent processing method and system in the technical field of intelligent operation and maintenance, and the method comprises the steps: S10, obtaining a large amount of work order data, and carrying out the data cleaning of each piece of work order data, and obtaining a work order data set; step S20, creating a work order pre-judgment model, a work order transfer model and a work order reply model, and training each model by using the work order data set; step S30, issuing the trained work order pre-judgment model, work order transfer model and work order reply model; step S40, acquiring a to-be-processed work order, performing data cleaning on the to-be-processed work order, inputting the to-be-processed work order into the work order pre-judgment model to perform work order classification, and generating a classification result; step S50, inputting the to-be-processed work order into a work order transfer model based on the classification result so as to transfer the to-be-processed work order; and step S60, intelligently processing the work order to be processed through the work order reply model. The method has the advantage that the work order processing efficiency is greatly improved.

Description

technical field [0001] The invention relates to the technical field of intelligent operation and maintenance, in particular to a method and system for intelligently processing operator work orders based on NLP. Background technique [0002] With the continuous increase of information, new challenges have been brought to the operation and maintenance work, especially the operation and maintenance of operators. After receiving the complained work order, the operator's customer service personnel need to forward the work order to the corresponding operation and maintenance personnel. The operation and maintenance personnel will judge the validity of the work order, return the invalid work order, dispatch the valid work order, and Manually query the troubleshooting manual. However, the operation and maintenance of traditional operators has the following problems: [0003] 1. Customer service personnel receive tens of thousands of complaint work orders every day, and there are v...

AI Technical Summary

Problems solved by technology

[0003] 1. Customer service personnel receive tens of thousands of complaint work orders every day, and there are various types of problems. Different types of problems require different technical personnel to assist in handling them; 2. Operation and maintenance personnel need to judge the validity of work orders and return invalid work orders , the customer service personnel need to re-fill the work order and submit it again, which increases the manual input of work order preprocessing and generates a large amount of invalid workload, resulting in prolonged work order processing time and low work order processing efficiency; 3. Operation and maintenance personnel based on their own experience Sending orders, wrong dispatching orders will lead to problems that cannot be resolved, and need to be returned for re-dispatching orders, which increases the work order processing process and the invalid processing time of operation and maintenance personnel, and reduces the efficiency of work order processing; 4. Manually query faults Handling manuals for question comparison, for scenario questions with a high monthly frequency, repeated queries and answers are required, increasing the workload of repetition

Images