A SYSTEM THAT PROVIDES AUTOMATIC MANAGEMENT OF CUSTOMER COMPLAINTS.

Abstract

This invention relates to a system (1) developed to ensure customer satisfaction by interpreting and classifying customer sentiment in complaint calls made to telecommunication companies.

Description

1 TARIFF PROVIDING AUTOMATIC MANAGEMENT OF CUSTOMER COMPLAINTS A SYSTEM Technical Area This invention addresses customer complaints made to telecommunications companies. a system that ensures customer satisfaction by interpreting and classifying feelings It is related to the system. 10 Previous Technique In the current state of the art, a wide variety of workforces typically utilize software applications and... The tools are supported by business intelligence (BI) tools. In general, the call is for the big 15 Companies, agencies, institutions, and other organizations process large amounts of data (for example, a to process and / or accurately record the transcript (data) between the customer and a representative They are constantly facing a problem with stopping. At the store level. or efficient and rapid planning of activities at the market / region level is important It offers. Traditional tools only require the manual operation of complaints. 20 by preventing unnecessary transmission to higher levels and resulting in low quality and performance. This can lead to problems. Therefore, considering the studies and shortcomings in the current technique... When this is taken into account, 25% of complaint calls made to telecommunications companies Text and sentiment analysis is performed according to complexity and requirements. It appears that a system is needed to enable classification. 2 United States Regulation US2023197105A1, which is included in the known state of the art. in the patent document, the call or conversation between the customer representative and the customer a real-time analysis of call log data generated during the process The system in question is described. The system in question, as described in the invention, contains call transcript data. By analyzing the real-time data stream, dissatisfaction data can be automatically generated. 5 It creates a data definition module that identifies dissatisfaction as such. The system is between an application and a cloud platform deployed in a cloud environment. It communicates with each other. The application uses a service layer module, an interface layer module, and It includes the user interface (UI) layer module. The service layer module, Call transcript created during a call or chat between the customer and the representative 10 It receives a real-time stream of data. Then, call transcript data. It is applied to a machine learning model and with predefined complaint data. They are compared. This comparison leads to the creation of a first similarity score. It provides and the call depends on whether this score is greater than a certain threshold value. Transcript data is automatically processed as a primary dissatisfaction data point. 15 It is being defined. A second set of dissatisfaction data is being automatically identified. And when new call transcript data is received, a second similarity score is automatically calculated. This involves training a machine learning model to define it as such. The system, The agency can receive feedback on complaint data from its computer device, can store the data in a database and an analyst 20 that can be used by analysts. can be consumed by the computing device. This includes voice or text data. voice data from the telephone device covering various situations in which it can be used While text data can be retrieved from elsewhere, text data can be retrieved from the chat platform. The system also... We can generate suggestion data and use this to solve problems raised by customers. It can provide data-driven recommendations. This recommendation data includes complaint data, policy 25 The data can include various information such as complaint types and subtypes. To generate and display recommendation data based on dissatisfaction data. The UI interface layer module can call other modules used in the system. 3 the components of the machine learning model, especially natural language processing (NLP) Language Processing) model and long short-term memory (LSTM) It includes models such as the (memory) model. Thus, call transcript data real-time generation of all statements between the customer and the representative combined to convert text into data and feed it into a machine learning model 5 by implementing it, dissatisfaction data that emerges during the call is quickly obtained. identification is provided. Brief Description of the Invention The purpose of this invention is to process customer complaints made to telecommunications companies in text format. and by using technologies such as sentiment analysis, the cause of complaints, emotional by determining its tone and priority, in order to ensure it is resolved quickly. It is about implementing a developed system. Another purpose of this invention is to support machine learning models and rule-based processing. with automated actions supported by workflows, the complaint management process Automating the classification, prioritization, and resolution of complaints developed to enable processes to be made more automated and data-driven It is about creating a system. 20 Detailed Description of the Invention The "Automated Customer Complaint Processing" system was developed to achieve the purpose of this invention. A system that provides management is shown in the attached figure; this figure is 25 Figure 1 shows a schematic view of the system that is the subject of the invention. 4 The parts shown in the figure are individually numbered, and the corresponding numbers correspond to these numbers. It is given below. 1. System 2. Electronic Device 5 3. Database 4. Analytics Server 5. Server Complaint calls made to telecommunications companies are processed quickly within 10 days. classification and prioritization, increasing customer satisfaction and service by enabling the management of flows subject to the level agreement, customer The invention was developed to enable the automatic classification of complaints. subject system (1); - communication with the customer via a GSM operator in written, audio or visual form 15 at least one electronic device configured to enable it to be installed (2), - customer history, product information, etc. via electronic device (2) first, whether they have experienced a similar problem before, the problem resolution time, previous experience If a grade was given in the complaint, the grade point average and the number of referrals the complaint was resolved through are also included. By storing information, it performs data retrieval and preparation processes using ETL operations and 20 at least one database structured to provide evaluation (3), - Analytics on data obtained and kept under record via database (3) by conducting studies, performing tasks such as classification, prioritization, and sentiment analysis. by implementing algorithms for text analysis and sentiment analysis, complaints at least one analysis server configured for classification (4), 25 - Complaints classified by being in communication with the analysis server (4) and database (3) defining rule-based process flows of calls, machine learning using models to create complex sequences of operations and automate actions. at least one server configured to ensure customer satisfaction by performing this task (5) includes. The electronic device (2) in the system (1) which is the subject of the invention enables data entry. 5 that allows for the use of keys or a touchscreen and the detection of ambient sound data. a smartphone, tablet computer or portable computer that has a microphone is a device. The electronic device in question (2) does not use any remote communication protocol. by establishing a connection with the server (5) and through this established connection, the database to enable data exchange between (3) and the server (5) It is being structured. 10 The database (3) in the system (1) which is the subject of the invention, via the electronic device (2) The customer's product information, previous problems in a similar category if they didn't experience it, if they gave a grade in their previous complaints, what is their grade point average, how many referrals? The complaint has been resolved and the customer is notified of the problem resolution time by storing the ETL (Extract, 15 Data retrieval and processing through Transform and Load operations. to carry out the preparation processes and provide evaluation It is structured. The database in question (3) handles complaint calls with attributes. and is structured to be prepared according to metrics. In the system that is the subject of the invention, the analysis server (4) located in (1) is in communication with the server (5). and is configured to be managed by the server (5). The invention is preferred In the arrangement made, the analysis server (4) receives data from the database (3) Term Frequency-Inverse on data obtained from customer phone calls Text analysis using Document Frequency (Term Frequency - Reverse Document Frequency) 25 It is configured to do so. The analysis server (4) stores the analyzed texts in Naive To classify customer complaints using the Bayes (Pure Bayes) algorithm. It is being structured. 6 The server (5) in the system (1) which is the subject of the invention, any remote communication to communicate with the electronic device (2) using the protocol and the established communication data exchange with the analysis server (4) with the data received from the database (3). It is configured to perform. The server (5) inserts new data into the database (3). data recording, deletion of recorded data in database (3) or data 5 Modifying the registered data in the database (3) and the data in the database (3) to manage the database (3) by operations such as updating the registered data It is structured. The server in question (5) is structured according to the rules for classified complaint calls. It is configured to define the process flows based on the server (5), using machine learning models to generate complex sequences of operations and 10 by performing automated actions to ensure customer satisfaction It is being structured. Industrial Applicability Thanks to the system (1) that is the subject of the invention, telecommunication companies can address customer complaints. It collects data from calls and uses text analysis and sentiment analysis modules. using complaints to classify, prioritize, and process them based on rules. By defining workflows, efficiency is achieved in operational processes. Around these fundamental concepts, the invention focuses on "Automated Customer Complaint Handling". Development of a wide variety of applications related to "A System that Provides Management (1)" It is possible, and the invention is not limited to the examples described here, but is essentially As stated in the requests.

Claims

7 REQUESTS 1. Complaint calls made to telecommunication companies should be processed quickly. classification and prioritization, increasing customer satisfaction and by enabling the management of flows subject to service level agreements 5 It enables the automatic classification of customer complaints; - the customer receives written, audio or visual communication through a GSM operator. at least one electronic device configured to enable communication (2), - Customer history and product information via electronic device (2), Whether they have previously experienced a similar problem in this category, problem resolution time, 10 If they gave a grade in their previous complaints, what is their grade point average and how many referrals were made? By storing information about when the complaint has been resolved, it retrieves data through ETL operations and... to carry out the preparation processes and provide evaluation containing at least one structured database (3) and - data received and kept under record via database (3) 15 Classification, prioritization, and sentiment analysis through analytical studies. algorithms for text analysis and sentiment analysis by performing operations such as these. to include at least one analysis structured to classify complaints. server (4), - communicating with the analysis server (4) and database (3) and classified 20 Defining rule-based process flows for complaint calls, machine to create complex sequences of operations using learning models and by performing automated actions to ensure customer satisfaction a system characterized by having at least one server (5) configured (1).

2. Keypad or touchscreen for data entry and ambient sound. smartphones and tablets equipped with microphones that enable data detection 8 electronic device (2) characterized by a computer or portable computer A system like the one in Request 1 (1).

3. Connect to the server (5) using any remote communication protocol. to establish and connect the database (3) and the server (5) through this established connection. an electronic device configured to enable data exchange A system like the one in Claim 1 or 2 characterized by (2) (1).

4. Product information owned by the customer via electronic device (2), more first, whether they have experienced a similar problem before, in their previous complaints 10 If a grade was given, the grade point average, how many referrals were made to resolve the complaint, and Retrieving data through ETL transactions by storing the problem resolution time for the customer and to carry out the preparation processes and provide evaluation from the above requests characterized by the structured database (3) a system like any other (1). 15 5. Structured to prepare complaint calls according to features and criteria. any of the above requests characterized by the database (3) such a system (1).

6. To be in communication with Server (5) and to be managed by Server (5) from the above requests characterized by the configured analysis server (4) a system like any other (1).

7. Data obtained from the database (3) and 25 from the customer's phone calls Text based on Term Frequency-Inverse Document Frequency in data. characterized by the analysis server (4) configured to perform the analysis a system like any of the above requests (1). 9 8. Analyzing the texts to identify customer complaints using the Naive Bayes algorithm. characterized by the analysis server (4) configured for classification a system like any of the above requests (1).

9. Using any remote communication protocol, contact electronic device (2) with 5 to communicate and to obtain from the database (3) through this communication to exchange data with the analysis server (4) any of the above requests characterized by the configured server (5) a system like one of them (1).

10. Making new data entries into the database (3), within the database (3) deletion of registered data or registered data in the database (3) modification and updating of registered data in the database (3) Server (5) configured to manage the database (3) with operations such as A system like any of the above-mentioned systems characterized by 15 (1).

11. Rule-based process flows for classified complaint calls. The above is characterized by the server (5) configured to define it. a system like any of the requests (1). 20 12. Generating complex sequences of operations using machine learning models. and ensure customer satisfaction by performing automated actions the above requests are characterized by the server (5) configured to be so. a system like any other (1). 25

Images