A system that estimates failure termination time using Bayesian non-negative matrix factorization (BNMF) and communicates this to the user via a chatbot.

Abstract

This invention relates to a system (1) developed to predict the termination times of current alarms by analyzing the historical alarm termination data of internet connection devices with Bayesian Non-Negative Matrix Factorization (BNMF) algorithms, and to show the user the estimated termination time of the fault alarm through the relevant interface when the subscriber using the internet connection device enters that there is a problem with his connection through an interface.

Description

1 TARIFF FAILURE TERMINATION TIME BAYESIAN NON-NEGATIVE MATRIX PREDICTING AND CHAT ROBOT USING FACTORIZATION (BNMF) A SYSTEM THAT TRANSMITS INFORMATION TO THE USER VIA A NETWORK Technical Area This invention enables the processing of historical alarm termination data from internet-connected devices using Bayesian technology. By analyzing it using Non-Negative Matrix Factorization (BNMF) algorithms. Predicting the expiration times of current alarms, internet connection device 10 The subscriber is experiencing a problem connecting to a chatbot interface. The estimated time for the fault alarm to end is determined by the actions taken as a result of entering the information. A system that allows the duration to be displayed to the user via the relevant chatbot. It is related to. Previous Technique Today, internet service providers offer their customers quality internet service. To be able to provide this, the connected devices must be kept constantly active and operational. They want this. The devices owned by internet service providers have various features in different situations. Malfunctions may occur. These malfunctions should be resolved as soon as possible to ensure uninterrupted service. Services must be provided. Therefore, historical alarm termination data of internet connection devices are Bayesian. By analyzing with Non-Negative Matrix Factorization (BNMF) algorithms, 25 Estimating the expiration times of current alarms, internet connection device. The subscriber is experiencing a problem connecting to a chatbot interface. The estimated time for the fault alarm to end is determined by the actions taken as a result of entering the information. 2 a system that allows the duration to be shown to the user via the relevant chatbot It is understood that it is needed. International patent number WO2023164295, which falls under the prior art. The document describes cellular 5 in a wireless communication network such as a wireless 5G network. systems for managing user experience during communication interruptions and The methods are described. Example configurations include network communication error testing. managing user experience during cellular network outages It includes systems and methods for, specific telecommunication network components and Response to identifying errors in communication between specific network connection errors 10 various rescue actions are carried out, including radio communication in the distributed unit. during a cellular telecommunications network outage involving the resource manager Managing user control, future communication with distributed unit (DU). cellular system containing an intelligent layer module that makes predictions about communication failures Managing user experience during telecommunications network failures / outages, 15 Cellular telecommunications network disruption by utilizing a backup data center. In this context, user experience management, cellular telecommunications network A new central hub will be added to manage user experience during outages. Central units (CUs) will be resized to allow for units. This involves the inclusion of a corresponding group of microservices. 20 Brief Description of the Invention The purpose of this invention is to analyze historical alarm termination data from internet-connected devices. Analysis with Bayesian Non-Negative Matrix Factorization (BNMF) algorithms 25 By doing this, it is possible to estimate the expiration times of current alarms, internet connection when a subscriber using their device connects to a chatbot interface The problem was identified, and the resulting actions terminated the fault alarm. 3 Enable the estimated time to be displayed to the user via the relevant chatbot. The goal is to implement a system developed for this purpose. Detailed Description of the Invention The “Failure Termination Time Bayesian Non-Bayesian” technology was developed to achieve the purpose of this invention. Predicting with Negative Matrix Factorization (BNMF) and via Chatbot A "System that Delivers Information to the User" is shown in the attached diagram; Figure 1; Schematic view of the system that is the subject of the invention. 10 The parts shown in the figure are individually numbered, and the corresponding numbers correspond to these numbers. It is given below. 1. System 15 2. Chatbot Application 3. Database 4. Server Historical alarm termination data of internet connection devices Bayesian Non-20 Current analysis using Negative Matrix Factorization (BNMF) algorithms. estimating alarm termination times using an internet-connected device The subscriber is having trouble connecting to a chatbot interface. The estimated time for the fault alarm to terminate as a result of the actions taken upon its entry. 25 developed to ensure that the relevant chatbot is shown to the user. The subject of the invention is the system (1); -to enable internet service users to access fault notifications at least one chatbot application configured (2), 4 - By contacting the company that provides internet access services, user data is recorded. at least one database structured to keep it under (3), -Check the alarm information to determine if the alarm is for a planned outage or a known fault. to investigate whether or not there is a planned outage and / or known fault through investigation When it is determined that there is a disruption and / or 5 teams on the database (2) access the estimated completion time information reported by the chatbot application for the outage. (2) at least one configured to provide the user with information. Includes server (4). The chatbot application (2) included in the system (1) which is the subject of the invention, any communication 10 It is configured to communicate with the server (4) using the protocol. Chat robot application (2) sends data to server (4) using any communication protocol. It is structured to carry out the shopping. Chatbot application (2), The user completes the login process through an interface to resolve the issue. It is structured to enable users to request information. Chatbot 15 application (2), internet service users have a fault on an interface It is structured to provide them with information. Chatbot application (2), users can apply through any electronic device It is structured to enable them to access it. Chatbot application (2), 20 allows the user to view information and initiate the alarm inquiry process. It is structured to enable the user to obtain information. (chatbot application) (2) the server (4) queries whether there is an active failure on the device on the node After this is done, if there is no active alarm, a recorded alarm will be created. the user should be informed that there is no such thing and if there is an active alarm. The fault details and estimated completion time information returned from the flow are provided to the user in a 25 It is configured to allow transmission via the interface. The database (3) in the system (1) which is the subject of the invention, any communication any It is configured to communicate with the server (4) using the communication protocol. The database (3) exchanges data with the server (4) using any communication protocol. It is structured to carry out the database (3), internet provider company The database is structured to keep customer data on record (3), 5 The customer receives information about the first node to which the customer modem is connected (node ​​info). It is structured to keep records based on the number. Database (3), the data sent by the device on the node to which the internet provider's customers are connected It is configured to keep a record of the last active alarm information. Data base (3), estimated termination time data given by teams for outages or malfunctions 10 It is structured to keep records. In the system that is the subject of the invention (1), the server (4) located in any communication any Communicating with the chatbot (2) and database (3) using the communication protocol It is configured to do so. The server (4) can use any communication protocol 15 to exchange data with the chatbot (2) and the database (3) It is configured. The server (4) checks the alarm information and the alarm is scheduled. to inquire whether it relates to an interruption or a known fault It is being configured. The server (4) received the alarm as a result of the inquiry. If the outage occurred due to a planned outage or a known fault, the teams from the database (3) reported the outage 20 or retrieve the estimated completion time they provided for the malfunction and respond to the user. It is configured to ensure the time is returned. The server (4) is configured to ensure the time is returned. The investigation concluded that the alarm in question was not caused by a planned outage or a known fault. and the number of accidents that occurred in the last year on the device on the node to which the customer is connected Retrieve the list of alarms from the database, calculate the alarm termination time, and 25 to provide the user with the calculated response time is being configured. The server (4) determines the planned outage and / or as a result of the query. When it is determined that there is no known fault, the relevant subscriber is connected from the database (3). 6 a list of alarms that occurred on the modem at that node in the last year access and data on alarms that occurred within the last year in a data-time format. It is configured to model. The server (4) models the alarm based on time data. by normalizing the data and preferably creating 1-hour periods It is configured. The server (4) displays the values ​​in 1-hour periods where no data is found. It is configured to perform interpolation in detection. Server (4), Iteratively applies BNMF to modem alarm values ​​recorded hourly. by running the algorithm, to separate the data-time matrix into two distinct matrices and To create a data pattern, analyze the dataset collected up to time point t using time t+1. Expanding the point so that it becomes 0, thus obtaining 2 10 from the initial matrix. The row corresponding to the point t+1, obtained by multiplying the separate matrices, is the modem of the relevant subscriber. to ensure it includes the estimated alarm termination time value It is configured. Server (4) generates the estimated alarm with the BNMF algorithm. the expiration time via chatbot (2) through the notification application It is configured to ensure that the user is notified. 15 Industrial Applicability Thanks to the system (1) that is the subject of the invention, the historical alarm of internet connection devices termination data Bayesian Non-Negative Matrix Factorization (BNMF) 20 By analyzing current alarm termination times with algorithms The prediction is that the subscriber using the internet connection device will access the site via an interface. The fault alarm was triggered as a result of the actions taken after entering the connection problem. The estimated completion time is displayed to the user via the relevant interface. is provided. 25 Based on these fundamental concepts, the invention focuses on "Failure Termination Time Bayesian Non-Bayesian". Predicting with Negative Matrix Factorization (BNMF) and via Chatbot 7 Development of a wide variety of applications related to "A System that Delivers to the User (1)" It is possible, and the invention is not limited to the examples described here, but is essentially As stated in the requests.

Claims

8 REQUESTS 1. Bayesian analysis of historical alarm termination data of internet connection devices. Analysis with Non-Negative Matrix Factorization (BNMF) algorithms By doing this, it is possible to estimate the expiration times of current alarms, internet 5 The subscriber using the connection device has a problem connecting via an interface. The fault alarm ends as a result of the actions taken upon its entry. enabling the forecast period to be displayed to the user via the relevant interface; - enabling internet service users to access fault notifications including at least one chatbot application (2) configured to provide and 10 -by contacting the company that provides internet access service, the user at least one database structured to keep its data on record (3), -Check the alarm information to determine if the alarm is for a planned outage or a known malfunction. to inquire whether or not there is a planned outage, and / or through this inquiry When the known fault is detected, 15 located on the database (2) information on the estimated completion time reported by the teams for the outage and / or malfunction access and inform the user via the chatbot application (2) a server (4) characterized by having at least one server configured to provide system (1).

2. To communicate with the server (4) using any communication protocol. In Claim 1, characterized by the configured chatbot application (2) such a system (1).

3. Exchange data with the server (4) using any communication protocol 25 chatbot application (2) configured to perform A system like the one in Claim 1 or 2, characterized (1). 9 4. The user performs login procedures through an interface to troubleshoot the problem. chat structured to allow request for termination information any of the above requests characterized by the robot application (2) a system like one of them (1).

5. Internet service users receive fault notifications via an interface. chatbot application (2) configured to enable them to receive a system like any of the above characterized demands (1).

6. Users can access the application from any electronic device. chatbot application (2) configured to enable them to access a system like any of the above characterized demands (1).

7. The user can view the information and the alarm inquiry process has begun. chatbot application structured to enable it to obtain information 15 (2) like any of the above-mentioned claims characterized by system (1).

8. The server (4) queries whether there is an active fault on the device on the node. After this is done, if there is no active alarm, a registered 20 The user should be informed that there is no alarm and an active alarm should be activated. In this case, the fault details and estimated termination time returned from the flow. to ensure that the information is transmitted to the user via an interface the above is characterized by the structured chatbot application (2) a system like any of the requests (1). 25 9. Any communication with the server (4) using any communication protocol characterized by a database structured for communication (3) a system like any of the above requests (1).

10. Exchange data with the server (4) using any communication protocol. characterized by the database structured to perform (3) a system like any of the above requests (1).

11. Internet service providers are required to keep customer data on record. 5 from the above requests characterized by the structured database (3) a system like any other (1).

12. Information about the first node (node ​​info) to which the customer modem is connected. Data structured for keeping records based on customer number 10 as in any of the above claims characterized by base (3) a system (1).

13. The device on the node to which the internet provider's customers are connected. 15 to keep a record of the last active alarm information data it sent. from the above requests characterized by the structured database (3) a system like any other (1).

14. Estimated completion time data provided by the teams for outages or malfunctions. 20 characterized by a database structured to keep records (3) a system like any of the above requests (1).

15. Any communication using a chatbot with any communication protocol. (2) and the server (4) configured to communicate with the database (3) A system like any of the above-mentioned demands characterized (1). 25 16. Chatbot (2) and database (3) using any communication protocol Characterized by the server (4) configured to exchange data with a system like any of the above-mentioned requests (1). 11 17. Checking alarm information to determine if the alarm is related to a planned outage or a known event. with the server (4) configured to check whether it is related to the fault a system like any of the above characterized demands (1).

18. As a result of the investigation, it was determined that the alarm in question was either a planned outage or a known 5 If it is caused by a malfunction, the teams from the database (3) for the interruption or malfunction to retrieve their estimated completion time and respond to the user characterized by the server (4) configured to ensure the return of time a system like any of the above-mentioned requests (1).

19. As a result of the investigation, it was determined that the alarm in question was either a planned outage or a known one. If the malfunction did not occur, the last one on the device at the node to which the customer is connected. To retrieve a list of alarms that occurred within a year from the database and to alarm calculate the termination time and provide the user with the calculated time as a response. 15 characterized by the server (4) configured to enable the return. a system like any of the above requests (1).

20. The investigation revealed that there was no planned outage and / or known fault. when done, from the database (3) on the node to which the relevant subscriber is connected Access the list of alarms that occurred on the modem in the last year and the 20 in question. To create a data-time model of alarm data generated over the past year. the above requests are characterized by the server (4) configured to be so. a system like any other (1).

21. Normalize the data in the time-based alarm model, preferably 1-25. Characterized by the server (4) configured to create hourly periods a system like any of the above-mentioned requests (1). 12 22. Determining values ​​in 1-hour periods for which no data is available. characterized by the server (4) configured to perform interpolation a system like any of the above requests (1).

23. Iteratively analyze the modem alarm values ​​recorded hourly for 5 seconds. Separating the data-time matrix into two separate matrices by running the BNMF algorithm. to provide and create the data pattern, data collected up to time point t Expanding the time point t+1 in the set so that it becomes 0, thus starting the process. the point t+1 is obtained by multiplying two separate matrices derived from the matrix. The line should set the estimated alarm termination time value of the relevant subscriber's modem to 10. characterized by the server (4) configured to ensure that it includes a system like any of the above requests (1).

24. Information on the estimated alarm termination time generated by the BNMF algorithm. notify the user via chatbot (2) through the application 15 The above is characterized by the server (4) configured to provide a system like any of the requests (1).