A text clustering and analysis method for repeated customer service calls
A technology for customer service calls and text clustering, applied in text database clustering/classification, metadata text retrieval, unstructured text data retrieval, etc. Helps to make timely decisions, improve work speed, and improve work efficiency
Active Publication Date: 2019-04-09
国网山东省电力公司营销服务中心(计量中心) +2
View PDF3 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0003] At present, there are many problems in the analysis of repeated work orders: the content of work order acceptance is text information, which is difficult to analyze, and the amount of information is large and unstructured, requiring manual review one by one, which is cumbersome and inefficient
There is little analysis of repeated calls, and it is impossible to understand the reasons for repeated calls from customers in a timely manner, resulting in customer service lag problems, so that it is impossible to respond to hot issues in a timely manner
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0034] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.
[0035] 1 Repeat call concept
[0036] Repeated calls refer to the set of events in which the same user (with the same account number) calls twice or more for the same event. The specific description is as follows: In the query period, except for praise, subscription and one-time settlement consultation, all non-user number calls (such as 12345, internal dial test, etc.) The number of repeated events, the number of work orders, and the number of phone calls obtained through repeated screening of content, acceptance content, etc.
[0037] The customer number is used as the unique identification attribute to determine the customer, but in the actual call, most of the work order records are missing. However, the same customer may use different phone numbers to call, but it reflects the same thing. The problem reported by the customer's call can be determ...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The present invention discloses a text clustering and analysis method for repeating caller work orders of customer service calls. The method comprises the following steps: cleaning an extracted customer service work order, and removing an abnormal caller number among repeating calls; extracting work orders of the same calls by using an association rule, and establishing a repeating caller work order text set; establishing a vector space model, using a feature vector to represent a feature text, using a vector whose components are weights of feature items to represent an entire text, and determining similarity between texts; extracting repeating caller work orders with similar content from the same user through hierarchical clustering, and analyzing a semantic feature; and recording, in a timely manner, problems reflected by repeating caller work orders to form a list of problems reflected by the repeating caller work orders. According to the method, manual retrieval of the repeating caller work orders is replaced, thereby improving working efficiency.
Description
technical field [0001] The invention relates to a method for text clustering and analysis of customer service calls for repeated calls. Background technique [0002] In recent years, with the rapid development of the power business, the number of customers and business types have increased accordingly. The content of customer calls is complex and diverse. How to find useful information from the huge amount of traffic is a huge challenge for customer service. According to the analysis of repeated call information, the semantic features hidden in the call information are mined, focusing on customers with a large number of repeated calls, and timely discovering and tracking hot issues in calls, so as to grasp the best time to deal with the problem, thereby improving Ability to deal with hotspot issues and monitoring capabilities. Therefore, how to collect and discover customers' repeated calls in the call information, semantically analyze the repeated calls, and accurately gra...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06F16/38G06F16/35
CPCG06F16/3347G06F16/35
Inventor 刘勇超田晓李静于相洁王婧谭永梅杜颖刘宏国孟巍张春秋
Owner 国网山东省电力公司营销服务中心(计量中心)