Deep learning-based electric power suspected complaint work order recognition method

Abstract

The invention discloses a deep learning-based electric power suspected complaint work order recognition method, and relates to complaint recognition method for electric power customer services. At present, electric power suspected complaint work orders are artificially recognized, so that the efficiency is low, the timeliness is bad and human resources are wasted. The method comprises the following steps of: 1) deep learning model configuration; 2) complaint feature label extraction; 3) complaint sample formatting; 4) model learning and training; 5) suspected complaint recognition; and 6) suspected complaint classification. According to the method, a series of work such as data cleaning and sorting, complaint tendency word extraction, data modeling and optimization, sample iterative learning and training, machine learning and prediction and the like can be carried out on complaint lists through a deep learning technology, so that deep learning-based intelligent recognition and classification of suspected complaint work orders are realized, the intelligent work experience is enhanced and the service quality control work efficiency is improved.

Description

technical field [0001] The invention relates to a complaint identification method for electric power customer service, in particular to a method for identifying suspected electric power complaint work orders based on deep learning. Background technique [0002] From the analysis of 95598 incoming calls, although a large number of customers did not directly complain, or the agent misjudged that it was not a complaint, they expressed their dissatisfaction with the power supply service through consultation, opinions and suggestions. If it is not handled properly or not in time, it may be upgraded to a customer complaint. How to reduce the amount of user complaints and improve user satisfaction has become the focus of power supply enterprises. Effective analysis and management of complaints can improve customer satisfaction and loyalty. It is of great and far-reaching significance to continuously discover and improve the weak points of the power supply business, improve the serv...

AI Technical Summary

Problems solved by technology

[0003] At present, service improvement is inseparable from the analysis of customer demands, but traditional methods such as manual review of data and manual cleaning of data for index prediction have seriously failed to keep up with development needs, resulting in low efficiency, poor timeliness and waste of human resources

The province's annual call volume is as high as 8 million. The center adopts the traditional sample recording quality inspection mode, and listens to the recordings one by one manually. The work efficiency is low and it is impossible to accurately and efficiently extract customer complaints and dissatisfaction points, etc.

Only 3,456 man-hours are invested in the normal task of quality inspection for a single suspected complaint, and this will account for far less than 1% of the overall customer appeal mining work in the future

[0004] In addition, there are a certain number of complaints and wrongly dispatched work orders in the 95598 work order. The State Grid Marketing Department dispatched nearly 30,000 work orders from time to time, requiring the quality inspection to be completed within 6 days. The time is tight and the task is heavy. The center invested 6-7 People work overtime to get the job done

The workload of quality inspection is large, the efficiency is low, and the missed inspection rate is high. At the same time, due to the deviation of understanding of different quality inspection personnel, different results will result, and the quality inspection work cannot be carried out effectively.

Images