CALL CENTER MANAGEMENT SYSTEM WITH VIRTUAL TWIN SUPPORT

Abstract

This invention relates to an AI-powered system (1) that creates a virtual twin for the customer calling the call center, guiding the customer representative in resolving the customer's possible problems and requests.

Description

1 TARIFF CALL CENTER MANAGEMENT SYSTEM WITH VIRTUAL TWIN SUPPORT Technical Area This invention creates a virtual twin for the customer calling the call center, allowing the customer to... will guide the representative in resolving the customer's potential problems and requests. It is related to an artificial intelligence-powered system. Previous Tech 10 Current call center solutions require customer representatives to manually access customer data. This allows access to the system, but the process is time-consuming and prone to errors. This is possible. Furthermore, current systems often analyze historical data and... It fails to offer the right solution. For this reason, these problems 15 To resolve the issue, we instantly analyze all customer-related data and send it to the customer representative. a new approach that offers the most suitable solutions and provides guidance throughout the process The system is needed. Chinese patent CN116128520A, which is included in the prior art, 20 The document outlines how to create a digital twin of the customer and address their complaints more effectively. It refers to a system structured to ensure that the problem is solved in this way. One application of this invention is a customer management system based on digital twin technology. It describes the method and a related product. The method includes the following steps: target Obtaining target customer information belonging to the customer; digital twin technology 25 a digital twin of the target customer is created based on target customer information. the creation of the body; when the target customer reaches a target physical network point, Obtaining real-time customer data of the target customer; digital twin and A target customer's target behavior based on real-time customer data Predicting the intention; target management corresponding to the target behavior intention 30 Determining the strategy parameter; and the target management strategy parameter. 2 Performing a corresponding target operation. Implementing this invention. Thanks to this, customer behavior can be predicted in advance and customer complaints can be addressed. The efficiency of processing events is being increased. In one application of this document, Target customer information may include at least one of the following: customer age, gender, biological characteristics, occupation information, personality information, education information, 5 personal information, family relationships, preferences, account information, owned items (financial management, funds, precious metals held in the account, government bonds, insurance, term deposit certificates, etc.), public works, social network usage, internet behaviors, consumer behaviors, etc. There are no limitations within this scope. not available. 10 Brief Description of the Invention The aim of this invention is to create a virtual twin for the customer calling the call center, 15 Guiding the customer representative in resolving the customer's potential problems and requests The goal is to create an AI-powered system that will do this. Detailed Description of the Invention The "Virtual Twin Support Call 20" was carried out to achieve the purpose of this invention. The "Central Management System" is shown in the attached diagram; Figure 1. Schematic view of the system described in the invention. The parts shown in the figure are individually numbered, and the corresponding numbers are 25. given below: 1. System 2. Database 3. Server 30 3 A virtual twin is created for the customer calling the call center. will guide the representative in resolving the customer's potential problems and requests. artificial intelligence supported invention system (1); -Call history of customers calling the call center, product usage habits, Purchase behavior, customer feedback, resolution history, previous issues 5 and to ensure that data related to their solutions are kept on record in encrypted form. at least one database structured to (2) and -a virtual twin by analyzing customer data stored in the database (2) creating this virtual twin and guiding the customer representative through it. 10 by using artificial intelligence to offer the most suitable solution to customer problems guiding the customer representative throughout the call, past customer, past customer By reminding customers of their interactions, a more personalized service is offered, and the customer... configured to ensure data is securely stored in encrypted form It contains at least one server (3). In the system that is the subject of the invention, the database (2) is in communication with the server (3) (1). It is configured to exchange data with the server (3). In the preferred arrangement of the invention, speech content is included in the database (2), The customer's call details include dialogues with the representative, duration, and waiting times. all interaction data at the center, including the services and products used by the customer frequency of use, processing time, error log data, customer's previous solution The problems they are looking for, the product categories they purchase, and phone calls and mobile applications. recording data in the form of interaction preferences via messaging. It is structured to ensure that it is kept under control. The server (3) in the system (1) which is the subject of the invention communicates with the database (2) and is configured to access the data in the database (2). Server (3), conversation contents, dialogues with the representative, duration and waiting times this includes all interaction data of the customer with the call center, the customer's frequency of use of services and products, processing time, error log data, 30 problems the customer has previously sought solutions for, product categories they have purchased, and 4 Interaction in the form of phone calls and messaging via mobile applications. from call center servers where data in the form of preferences is integrated It is configured to ensure that the server (3) does not interrupt the call time. to filter out incomplete or missing recorded interviews before processing It is configured to detect and clean up erroneous data points. Server 5 (3), convert data formats such as date and time into a format that algorithms can understand. It is configured to convert the client's tone of voice. The server (3) new features as indicated in the call details in the form of sentiment analysis extraction and at this stage, using natural language processing (NLP) techniques It is structured to enable the extraction of meaning from speech content. 10 Server (3) classifies customer behaviors and acts according to these behaviors Using Support Vector Machines (SVM) to identify customer trends It is structured to handle customer complaints and information requests via SVM. Server (3) or customer past interaction data in the form of technical support requests It is structured to classify. Server (3), customer's call duration, 15 The data, such as tone of voice and type of problem, determines the decision classes of the SVM. to determine and ensure the most appropriate distinction is made between classes is being configured. The server (3) is configured when the customer makes a new support request. Naive Bayes to determine whether it is high stress or low stress It is configured to use the algorithm. Server (3), customer call 20 by comparing the history and tones of interaction with previous high-stress situations If the customer tends to experience high stress levels in certain areas, the customer becomes more sensitive. It is configured to direct the customer to representatives. The server (3) directs the customer to representatives. Offers to customers with similar behaviors, based on their previous interactions. K-Nearest 25 provides personalized recommendations by comparing solutions. It is configured to use the neighbor (KNN) algorithm. Server (3), With the KNN algorithm, when a customer encounters a new problem, it compares it to similar past data. By making predictions based on solutions offered to other customers, for the customer... It is configured to quickly predict the most likely solution. Server (3), the most suitable 30 among possible solution proposals to ensure customer satisfaction. using the random forest algorithm to determine which one it is It is configured in such a way. The server (3) will respond if the customer experiences technical problems. first, solutions offered to customers experiencing the same type of problem were presented in a random forest. It is structured to provide the best solution by analyzing it with its algorithm. Server (3) selects the highest possible solution from the possible solutions by using multiple decision trees. It is configured to determine which one has the success rate. Server (3), 5 Real-time status assessment based on the customer's experience during the call. do this, and if the customer is stressed, offer solutions that have previously proven highly successful. It is configured to activate the server (3), which has a high stress level. Customers are automatically assigned representatives who can better manage such situations. It is configured to assign. The server (3) assigns the customer's 10 during the call. Decision Trees algorithm shows appropriate solutions depending on the situation. It is configured to determine with the Server (3), from each customer interaction It is structured to train artificial intelligence models by feeding them with new data. Server (3) supervises the model with data obtained from real customer interactions. retraining is possible thanks to the supervised learning algorithm, and every new 15 It is structured to increase its capacity to make accurate predictions for the situation. Industrial Application of the Invention The system (1) that is the subject of the invention provides a virtual twin for the customer calling the call centre. by creating a system that enables customer service representatives to resolve potential customer problems and requests. Guidance is provided so that a quick solution is produced and customer satisfaction is ensured. is being increased. Around these fundamental concepts, the invention topic is “Call Center 25 with Virtual Twin Support”. It is possible to develop a wide variety of applications related to the Management System (1)” and the invention cannot be limited to the examples described here, but mainly to the claims as stated.

Claims

6 REQUESTS 1. A virtual twin is created for the customer calling the call center. guiding the representative in resolving the customer's potential problems and requests. will be AI-powered; 5 -Call history of customers calling the call center, product usage habits, purchasing behavior, customer feedback, solution History, data on previous problems and their solutions are encrypted. at least one database structured to ensure records are kept (2) and 10 -a virtual twin by analyzing customer data stored in the database (2) creating this virtual twin and guiding the customer representative through it. by offering the most suitable solution to customer problems, artificial intelligence By using this, the customer representative can be guided throughout the call and past By reminding the customer of past customer interactions, a more personalized 15 providing services and securely storing customer data in encrypted form. with at least one server (3) configured to ensure storage a characterized system (1).

2. Communicating with server (3) and exchanging data with server (3) 20 characterized by the database structured to perform (2) A system like the one in claim 1 (1).

3. Includes conversation content, dialogues with the representative, duration, and waiting time. All customer interactions with the call center in the form of times 25 data, frequency of use of services and products used by the customer, processing time, error log data, customer's previous attempts to resolve issues problems, product categories purchased, and phone calls, mobile application data in the form of interaction preferences via messaging database structured to ensure that records are kept (2) and 30 A system like the one in Claim 1 or 2, characterized (1). 7 4. To communicate with the database (2) and access the data in the database (2). the above characterized by the server (3) configured to access a system like any of the requests (1).

5. Conversation content, dialogues with the representative, duration, and waiting time. all customer interactions at the call center in the form of times data, frequency of use of services and products used by the customer, processing time, error log data, customer's previous attempts to resolve issues problems, product categories purchased, and phone calls, mobile application 10 data in the form of interaction preferences via messaging to ensure that it is retrieved from the call center servers with which it is integrated from the above requests characterized by the configured server (3) a system like any other (1).

6. Calls with interruptions or incomplete recordings will not be processed. by identifying and cleaning up missing or erroneous data points. from the above requests characterized by the server (3) configured for a system like any other (1).

7. Data formats such as date and time in a way that algorithms can understand. characterized by the server (3) configured to convert to form a system like any of the above requests (1).

8. Sentiment analysis in the form of 25, which determines the customer's tone of voice. The new features are being removed from the call details, and at this stage, natural language... using NLP (Neuro-Linguistic Programming) techniques to extract meaning from speech content Characterized by the server (3) configured to enable its extraction. a system like any of the above-mentioned requests (1). 8 9. Classifying customer behaviors and identifying customers based on these behaviors. Using Support Vector Machines (SVM) to identify trends from the above requests characterized by the server (3) configured for a system like any other (1).

10. SVM handles customer complaints, information requests, or technical support requests. in this way to classify customers' past interaction data from the above requests characterized by the configured server (3) a system like any other (1).

11. The customer's call duration, tone of voice, and type of problem The data allows SVM to determine decision classes and the best between classes. with the server (3) configured to enable the appropriate distinction to be made a system like any of the above characterized claims (1). 15 12. Is the customer experiencing high or low stress during a new support request? Using the Naive Bayes algorithm to determine if someone is stressed from the above requests characterized by the server (3) configured for a system like any other (1). 20 13. The customer's call history and interaction tones from previous high-stress periods. By comparing their situations, the customer experiences high stress levels regarding certain issues. If the tendency is to direct the customer to more responsive representatives 25 of the above requests characterized by the configured server (3). a system like any other (1).

14. Based on the customer's previous interactions, identify those who exhibit similar behaviors. By comparing solutions offered to customers, personalized recommendations are provided. 30 to use the K-Nearest Neighbor (KNN) algorithm to find 9 from the above requests characterized by the configured server (3) a system like any other (1).

15. When a customer encounters a new problem using the KNN algorithm, it compares it to similar past issues. Prediction based on solutions offered to other customers who have the data. to find and quickly estimate the most likely solution for the customer. from the above requests characterized by the configured server (3) a system like any other (1).

16. Select at least 10 possible solutions to ensure customer satisfaction. random forest to determine the most suitable one characterized by the server (3) configured to use the algorithm a system like any of the above-mentioned requests (1).

17. If a customer experiences technical problems, contact 15 other customers who have previously experienced the same type of problem. By analyzing the solutions offered to customers using the random forest algorithm. Characterized by the server (3) configured to offer the best solution a system like any of the above requests (1).

18. Using multiple decision trees, the highest success rate among possible solutions is determined. the server (3) configured to determine which one has the rate a system like any of the above characterized claims (1).

19. Current status according to the customer's experience during the call 25 We evaluate the situation, and if the customer is stressed, we can address the issue with a high success rate compared to previous years. Server configured to activate completed solutions (3) as in any of the above claims characterized by system (1). 10 20. For customers with high stress levels, these types of situations are better understood. configured to automatically assign representatives who can manage in any of the above requests characterized by the server (3) such a system (1).

21. Deciding on appropriate solutions based on the customer's situation during the call. structured to determine trees (Decision Trees) using the algorithm in any of the above requests characterized by the server (3) such a system (1).

22. AI models are fed with new data from every customer interaction. The above is characterized by the server (3) configured to train a system like any of the requests (1).

23. The model was supervised with data obtained from real customer interactions. 15 retraining thanks to the supervised learning algorithm and to increase the capacity to make accurate predictions for each new situation from the above requests characterized by the configured server (3) a system like any other (1). 25

Images