CN101320451A - Customer Complaint Approach Using Decision Trees - Google Patents

Abstract

The invention relates to a method for handling customer complaints based on a decision tree. The method for handling customer complaints is carried out according to the following steps: Step 1. Users submit complaints in various ways. Step 2. Process the complaints submitted by customers in a structured manner and extract structured information; Step 3. The rule engine queries the decision tree to trigger intelligent processing, and triggers the decision tree according to the structured information; Step 4. Automatically transfers to the background according to the definition of the decision tree Each intelligent processing unit conducts automatic query and analysis, and returns the analysis results to the customer service personnel or triggers the corresponding processing flow system; Step 5, converts the final processing results into customer language; Step 6, feeds back the processing results to customers in various ways The main function of the present invention is to automatically and accurately locate, analyze, and process customer complaints based on the decision tree, thereby reducing the technical threshold for customer complaint handling and promoting the ability to move forward in complaint handling.

Description

Customer Complaint Approach Using Decision Trees

technical field

The invention relates to a method for analyzing and solving customer complaints, in particular to a decision tree processing method, and belongs to the field of mobile communication.

Background technique

The traditional complaint handling system and some other auxiliary tools are difficult to meet the requirements of service management, the main reasons are as follows:

1. The original customer service support tools are many and complex, and they are isolated and unrelated, unable to form a unified whole, which not only has a great impact on the use of front desk customer service personnel, but also cannot give full play to the platform benefits of customer service support Investment returns.

2. The original customer service support tools are independent, irrelevant, and have a low degree of automation. They mainly rely on the personal experience of the processing personnel, and knowledge cannot be solidified and shared, so that the accuracy and timeliness of complaint handling cannot be further improved.

Contents of the invention

The purpose of the present invention is to provide a method for handling customer complaints based on a decision tree, which can realize automatic and accurate positioning, analysis, and processing of customer complaints, speed up the ability of complaint handling to move forward, and define a good decision tree process to perform tasks Automatic triggering, intelligent processing, and collaborative work across the network.

The customer complaint processing method based on decision tree of the present invention is carried out as follows:

Step 1. Users submit complaints in various ways.

Step 2. Process the complaints submitted by customers in a structured manner, and extract structured information (for example: complaint location, complaint business, complaint time, customer grade, etc.);

Step 3. The rule engine queries the decision tree to trigger intelligent processing, and triggers the decision tree according to the structured information;

Step 4. According to the definition of the decision tree, automatically go to the intelligent processing units in the background for automatic query and analysis, and return the analysis results to the customer service personnel or trigger the corresponding processing flow system;

Step 5, converting the final processing result into the customer's language;

Step 6. Feedback the processing results to the customer in various ways;

The system server includes core service equipment, data service bus-related equipment, and external systems. The core service equipment is composed of a complaint acceptance WEB server, a decision tree management rule engine server, and one-key query, analysis and positioning. , the background processing server that dispatches sheet processing is made up of; Described data service bus related equipment is made up of data service bus, interface adaptation server and API, the interface server of different types of webservice; Described external system is composed of short message, IVR system , Customer service work order, automatic dial test, alarm system, process system, knowledge management system, and customer complaint handling tool system.

The main function of the present invention is to automatically and accurately locate, analyze, and process customer complaints based on the decision tree, thereby reducing the technical threshold of customer complaint handling, promoting the ability of complaint handling to move forward, and triggering decision-making according to the structured processing of customer complaint information Tree judgment, automatic triggering of tasks, realizing collaborative work across the network, and further improving the quality and level of customer service.

The invention can effectively strengthen customer complaint response and support, realize the integration and fusion of complaint handling capabilities, form an intelligent complaint handling platform, and form an automatic and rapid response means for complaint positioning, analysis, and handling. The technical solution of the present invention will be described in further detail below with reference to the accompanying drawings.

Description of drawings

Fig. 1 is a core block diagram of the present invention;

Fig. 2 is a structural diagram of the system service composition of the present invention

Fig. 3 is a flow chart of method steps of the present invention;

Fig. 4 is a hardware structure diagram of an embodiment of the present invention;

Fig. 5 is the decision tree schematic diagram of core module of the present invention;

Fig. 6 is a structural diagram of each sub-module of the core module of the present invention.

Detailed ways

As shown in Figure 1, it is a core module diagram of the decision tree-based customer complaint handling method of the present invention, which mainly includes four parts: a complaint information analysis module, a decision tree definition module, an intelligent processing unit, and a rule engine.

Complaint information analysis module, for structural processing of customer complaint information, analyze the structured information content, as the input parameters for subsequent triggering decision tree process, such as: complaint phone number, complaint business type, complaint location, complaint time, etc.

The decision tree definition module mainly defines a set of follow-up processing links based on various input structured customer complaint information, defines in detail the intelligent processing unit called on each step, and performs subsequent process processing according to the results returned by the processing unit. Until the entire decision tree flow is completed.

The intelligent processing unit, as a link in the decision tree processing flow, mainly performs intelligent automatic processing according to the input structural parameters, and returns the processing results.

The rule engine, at the core of the entire functional module, is a high-performance special-purpose interpreter, which includes commands such as if... then how, which can analyze converted values and objects according to predefined rules, and then return Modified values and objects, or perform operations directly.

As shown in Figure 2, it is a system architecture diagram of the present invention, including the following parts:

1. Core service equipment: Complaint acceptance WEB server, decision tree management rule engine server, and background processing server that carries one-key query, analysis and positioning, and order processing.

2. Data service bus related equipment: data service bus, interface adaptation server, and different types of interface servers such as API and webservice.

3. External systems: including SMS, IVR and other systems, customer service work orders, automatic dial testing, alarm systems, process systems, knowledge management systems, customer complaint handling tools and other various systems.

The above three parts cooperate with each other to form the overall system architecture.

As shown in Figure 3, it is a processing flowchart of the present invention, comprising the following steps:

Step 1. Users submit complaints in various ways.

Step 2. Process the complaints submitted by customers in a structured manner, and extract structured information (for example: complaint location, complaint business, complaint time, customer grade, etc.);

Step 3. The rule engine queries the decision tree to trigger intelligent processing, and triggers the decision tree according to the structured information;

Step 4. According to the definition of the decision tree, automatically go to the intelligent processing units in the background for automatic query and analysis, and return the analysis results to the customer service personnel or trigger the corresponding processing flow system;

Step 5, converting the final processing result into the customer's language;

Step 6. Feedback the processing results to the customer in various ways;

Fig. 4 is the hardware structure diagram of the embodiment of the customer complaint processing method based on the decision tree of the present invention. In this embodiment, a commonly used three-layer architecture is adopted, which is mainly divided into a data layer, a data processing layer and an application service layer, and data storage is mainly placed in Performed on disk arrays to ensure high availability of data. The data processing layer mainly includes database servers and rule engine servers, all of which use enterprise-level servers. Application servers are mainly used to publish upper-layer applications. In the implementation process, it is necessary to consider avoiding single points of failure, and to expand services horizontally if necessary. In order to achieve high availability, a cluster is formed between two servers to achieve mutual backup. When one server fails, the application or database is automatically switched to another server.

As shown in FIG. 5 , it is a complaint handling decision tree of an embodiment of the method for handling customer complaints based on a decision tree in the present invention. In this embodiment, a rule engine is used to drive the decision tree. According to the complaint work orders received by the system, start the corresponding decision tree processing process for different complaint types. The information that needs to be queried is obtained at all levels of nodes in the decision tree, combined with the knowledge base for judgment, and finally achieves the goal of quickly locating problems and intelligently dispatching and processing them.

According to the rule-centered design concept, the present invention adopts the information object modeling technology, especially the information model automatic extraction technology, to abstract the information that needs to be exchanged, transferred, converted, merged, decomposed, etc. in the application to form an information model. It increases the maintainability of the entire software and prolongs the life cycle; at the same time, through the convenient modification of business rules, the response speed of the entire IT support is improved.

Figure 6 is a functional hierarchical structure diagram of an embodiment of the method for handling customer complaints based on a decision tree in the present invention. The system is divided into 10 modules such as complaint acceptance, analysis and positioning, rule management, complaint feedback, and decision tree management. The specific functions are shown in the following table Show:

The present invention needs to integrate a variety of different types of external interfaces, and the connection methods of each interface are different, so this method recommends adopting the industry-recognized enterprise information integration system architecture S0A service-oriented architecture method to analyze the functional requirements of each interface, Carry out service modeling, uniformly formulate service specifications, realize extracted services, and complete corresponding business processes based on services.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention without limitation. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be The scheme shall be modified or equivalently replaced without departing from the spirit and scope of the technical scheme of the present invention.

Claims (2)

1, a kind of customer complaining method that uses decision tree is characterized in that: be installed on the nucleus module in the system server, described nucleus module comprises the steps:

Step 1, user submit calling information to by variety of way;

Step 2, the calling information that the client is submitted to carry out structurized processing, extract structured message (for example: complain the place, complain business, complaint time, client's grade etc.);

Step 3, rules engine queries decision tree trigger Intelligent treatment, trigger decision tree according to structured message;

Step 4, each intelligent processing unit carries out the robotization inquiry and analyzes to the backstage automatically according to decision tree definition, and analysis result is returned the customer service treatment people or triggers the respective handling flow system;

Step 5, convert final process result to the client language;

Step 6, result is fed back to the client by variety of way;

Described system server comprises core service device, Data Service Bus relevant device, three part compositions of external system, described core service device is by complaint acceptation WEB server, decision tree rule engine server and the background process server composition that carries a key inquiry, analyzing and positioning, the single processing of group; Described Data Service Bus relevant device is made up of Data Service Bus, interface adaptation services device and the dissimilar interface server of API, webservice; Described external system is by note, IVR system, and customer service worker list, auto-dial testing, warning system, flow system, Knowledge Management System, customer complaint handling implement system form.

2, according to the customer complaining method of claims 1 described use decision tree, it is characterized in that: described nucleus module comprises calling information analysis module, decision tree definition module, intelligent processing unit module and rule engine module.

Images