An intelligent classification method of complaint work order based on convolution neural network

Abstract

The invention discloses an intelligent classification method of a complaint work order based on a convolution neural network. The invention comprises the following steps: step S1, collecting the textdata of the complaint work order, and constructing the text data set of the work order; step 2, performing word segmentation and cleaning on that text data of the complaint work order; 3, extract feature words and carry out quantization processing; step 4, train a convolution neural network to construct a classification model; Step S5: word segmentation and cleaning of the complaint work order text data to be classified and tested; Step S6: Converting the complaint work order text data to be classified and tested into vector representation; Step S7: Input the vector representation into the classification model. The invention realizes intelligent classification of the complaint work order by collecting the text data of the complaint work order in batch, cleaning and pretreating the data, training the classification model through the convolution neural network, inputting the vector representation into the analysis model, accelerating the processing speed of the complaint work order, andimproving the customer satisfaction and the service quality.

Description

technical field [0001] The invention belongs to the technical field of text classification, and in particular relates to an intelligent classification method for complaint work orders based on a convolutional neural network. Background technique [0002] In the era of Internet big data, real-time processing and analysis of user needs is particularly important. Taking the text processing of complaint work orders as an example, there are many types of user complaints. Operators can classify work orders by reading and understanding the content of complaint work orders, quickly identify user needs, and solve complaint problems. The timeliness of the complaint work order determines that the processing and distribution of the work order must be completed in a relatively short period of time. In this way, it is undoubtedly required that the classification of the complaint work order has a high real-time performance. [0003] The current method of classifying complaint work order t...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a method for intelligent classification of complaint work orders based on convolutional neural network, which generates work order text data by collecting complaint work orders in batches, performs data cleaning and preprocessing, and realizes the classification model through convolutional neural network Training, the vector representation of the text data of the complaint work order is input into the analysis model to realize the intelligent classification of the complaint work order, which solves the problems that the existing complaint work order is time-consuming, the customer service is easy to understand mistakes, and the work order processing efficiency is low.

Images