A system for user experience modelling for fixed wireless access service

Abstract

The present invention relates to a system (1) which enables customer satisfaction to be increased by performing precise quality of experience (QoE) assessment specific to fixed wireless access (FWA) users.

Description

[0001] A SYSTEM FOR USER EXPERIENCE MODELLING FOR FIXED WIRELESS ACCESS SERVICE

[0002] Technical Field

[0003] The present invention relates to a system which enables customer satisfaction to be increased by performing precise quality of experience (QoE) assessment specific to fixed wireless access (FWA) users.

[0004] Background of the Invention

[0005] In the state of art, quality of experience (QoE) measurement plays an important role in assessing user satisfaction, usability, performance and content quality across a variety of services, applications and technologies. Especially in recent years, quality of experience (QoE) measurement has been considered more important than quality of service (QoS). The methods used in known QoE measurement techniques are generally based on supporting a foundation formed by an objective blend of QoS data and content quality with machine learning. The common disadvantages of such methods are that they do not take into account the user types, and similarly, that they lack the ability to be adapted for different applications.

[0006] For this reason, considering the studies and deficiencies included in the current technique, it is understood that there is a need for a system which enables the relation between the quality of experience and quality of service of different types of users to become adaptive through fixed wireless access (FWA) user feedback.

[0007] The United States patent document no. US2023059500A1, an application included in the state of the art, discloses a system which is configured to extract experience and performance scores from experience data with machine learning method and to determine the actions to be taken in order to improve the customer experience. A method for automatically detecting and evaluating experience data associated with an experience journey to identify and adjust an action that is taken to improve customer experience is provided. In some embodiments, the method includes generating performance data associated with the experience journey from the experience data using a machine learning model. The method further includes determining the action to be taken based on analyzing the performance data. The method includes collecting new experience data responsive to the action having been taken and training the machine learning model using the new experience data. The method includes updating the performance data and the action to be taken based on training the machine learning model.

[0008] Summary of the Invention

[0009] An object of the present invention is to realize a system developed with the aim of enabling customer satisfaction to be increased by performing precise quality of experience (QoE) assessment specific to fixed wireless access (FWA) users.

[0010] Another object of the present invention is to realize a system developed with the aim of enabling a basic level user experience model, known as the customer experience index (CEI), to be developed; a quality of experience (QoE) to be created by using existing measurements, and therefore an innovative, low- complexity and highly efficient approach to be developed.

[0011] A further object of the present invention is to realize a system developed with the aim of enabling greater accuracy and consistency to be obtained; deviations in key performance indicators (KPIs) to be identified by taking user feedback into account, and therefore the ability to capture subtle differences in the user experience beyond what is perceived of the method to be developed by adapting the metrics to user perceptions. Detailed Description of the Invention

[0012] “A System for User Experience Modelling for Fixed Wireless Access Service” realized to fulfil the objectives of the present invention is shown in the figure attached, in which:

[0013] Figure-1 is a schematic view of the inventive system.

[0014] The components illustrated in the figure are individually numbered, where the numbers refer to the following:

[0015] 1. System

[0016] 2. Server

[0017] 3. Processor

[0018] E. Electronic Device

[0019] The inventive system (1) developed with the aim of enabling customer satisfaction to be increased by performing precise quality of experience (QoE) assessment specific to fixed wireless access (FWA) users comprises at least one server (2) which is configured to exchange data and to access the electronic device (E) by using any remote communication protocol; at least one processor (3) which is configured to establish connection with the server (2) by using any communication protocol; to output an overall satisfaction status by classifying complaining and non-complaining customers; to establish a relation between the probability of complaint and quality of experience (QoE) by using complaint data; output the Kolmogorov-Smirnov (KS) statistic between the discrepancy and cumulative distributions; to determine the appropriate CEI for each group; and to compare changes in performance measurements. The server (2) included in the inventive system (1) is configured to establish connection with the processor (3) by using any communication protocol included in the state of art. The server (2) is configured to access the electronic device (E) of the user / customer by using any communication protocol included in the state of art.

[0020] The processor (3) included in the inventive system (1) is configured to establish connection with the server (2) by using any communication protocol included in the state of art. The processor (3) is configured to use the customer experience index (CEI), a measurement calculated based on service performance indicators related to technology, terminal and application and consisting of four component categories as time, accessibility, quality and efficiency in order to measure quality of experience (QoE). The processor (3) is configured to classify each of the sub-metric measurements, expressed as a percentage representing the proportion of certain behavior of a user observed through a terminal and technology during application sessions, as “good”, “mediocre” or “poor”; to obtain the sub-metric values for each user, application, terminal and technology as a whole by calculating the percentage of “poor” examples in the total; and to obtain the experience index (CEI) by combining mainly these sub-metrics. The processor (3) is configured to collect small subscriber bases and to assign a group label of “other” to the resulting terminal CEI in order to handle cases where the number of subscribers per a terminal is low, and not to calculate the terminal or application CEI values, respectively, if the terminal could not be determined or if the given application has a small subscriber base. The processor (3) is configured to perform the classification of measurements into “good”, “moderate” and “poor” categories through the location in which their sub-metrics are located within the user community differentiated by the relevant application, terminal, and technology; to determine the percentage of “poor” examples after the labelling process; and to mainly collect these percentages for the groupings in the form of sub-metric, user, technology, terminal, and application in order to obtain CEI values related to the groupings in the form of user, technology, terminal, and application. The processor (3) is configured to model quality of experience (QoE) by using customer (user) feedback in the form of “complaints” and QoS (Quality of Service) oriented metric of CEI, and to divide customers or FWA (Fixed Wireless Access) subscribers into two groups as the customers who have not filed any complaints during a certain period of time and the others. The processor (3) is configured to develop a transition-to- complaint model for a quantitative quality of experience (QoE) in the form of a statistical QoE, threshold function, probability of complaint (continuous value) and complaint (binary outcome) by using current subscriber complaint data. The processor (3) is configured to assume that there is a probability of a user filing a complaint at any time depending on the hidden QoE values with the developed model. The processor (3) is configured to develop a model by handling complaint data collectively and focusing on whether a complaint has been filed or not during a certain period of time. The processor (3) is configured to provide an unknown threshold function in order to identify the relation between QoE and the probability of complaint. The processor (3) is configured to determine the probability of a complaint to occur by the proportion of cases where QoE does not exceed a certain threshold. The processor (3) is configured to predict the probability of complaint for a group of users by taking into account the proportion of data corresponding to users who have filed a complaint. The processor (3) is configured to measure the probability of a user complaining for the first time at a certain time by the probability that the QoE stays between two time-dependent thresholds. The processor (3) is configured to consider QoE not as time-varying but as a random variable that is measured in general. The processor (3) is configured to obtain the QoE model from the CEI by utilizing KPIs (Key Performance Indicators). The processor (3) is configured to measure the information in the form of time spent, total volume, round-trip time, efficiency, speed, and access time by using the KS (Kolmogorov-Smirnov) statistic between the relevant empirical Cumulative Distributions (CDFs) and the discrepancy between the complainants and the others. The processor (3) is configured to adjust the available CEI values corresponding to the respective percentiles, depending on the proportion of complaint data in each group, and to match them for QoE; and to consider the original CEI as a basis for QoE in order to be aligned with the previously mentioned model. The processor (3) is configured to perform the matching as adding an appropriate value to all available CEIs in one group and removing another appropriate value from all available CEIs in the other group, therefore to bring the corrected CEI values corresponding to the relevant percentiles to a common point and to enable this point to be decided by the CEI value in the percentile corresponding to the proportion of complaint data for all data. The processor (3) is configured to develop a measurement into which the CEI is converted, called a-CEI (Adjusted Customer Experience Index), as a representation of QoE and to decide on CEI adjustments by taking into account the change in a performance measurement that is determined to end.

[0021] Industrial Application of the Invention

[0022] In the inventive system (1), an overall satisfaction status by classifying complaining and non-complaining customers is output; a relation between the probability of complaint and quality of experience (QoE) is established by using complaint data; the Kolmogorov-Smirnov (KS) statistic between the discrepancy and cumulative distributions is output; the appropriate CEI for each group is determined; and the changes in performance measurements are compared. Therefore, it enable that a basic level user experience model, known as the customer experience index (CEI), is developed; a quality of experience (QoE) is created by using existing measurements, and therefore an innovative, low-complexity and highly efficient approach is developed.

[0023] Within these basic concepts; it is possible to develop various embodiments of the inventive “A System (1) for User Experience Modelling for Fixed Wireless Access Service”; the invention cannot be limited to examples disclosed herein and it is essentially according to claims.

Claims

CLAIMS1. A system (1) developed with the aim of enabling customer satisfaction to be increased by performing precise quality of experience (QoE) assessment specific to fixed wireless access (FWA) users; comprising at least one server (2) which is configured to exchange data and to access the electronic device (E) by using any remote communication protocol; and characterized by at least one processor (3) which is configured to establish connection with the server (2) by using any communication protocol; to output an overall satisfaction status by classifying complaining and non-complaining customers; to establish a relation between the probability of complaint and quality of experience by using complaint data; output the Kolmogorov- Smirnov statistic between the discrepancy and cumulative distributions; to determine the appropriate CEI for each group; and to compare changes in performance measurements.

2. A system (1) according to Claim 1; characterized by the server (2) which is configured to establish connection with the processor (3) by using any communication protocol.

3. A system (1) according to Claim 1 or 2; characterized by the server (2) which is configured to access the electronic device (E) of the user / customer by using any communication protocol.

4. A system (1) according to any one of the preceding claims; characterized by the processor (3) which is configured to establish connection with the server (2) by using any communication protocol.

5. A system (1) according to any one of the preceding claims; characterized by the processor (3) which is configured to use the customer experience index, ameasurement calculated based on service performance indicators related to technology, terminal and application and consisting of four component categories as time, accessibility, quality and efficiency in order to measure quality of experience.

6. A system (1) according to any one of the preceding claims; characterized by the processor (3) which is configured to classify each of the sub-metric measurements, expressed as a percentage representing the proportion of certain behavior of a user observed through a terminal and technology during application sessions, as “good”, “mediocre” or “poor”; to obtain the sub-metric values for each user, application, terminal and technology as a whole by calculating the percentage of “poor” examples in the total; and to obtain the experience index by combining mainly these sub-metrics.

7. A system (1) according to any one of the preceding claims; characterized by the processor (3) which is configured to collect small subscriber bases and to assign a group label of “other” to the resulting terminal CEI in order to handle cases where the number of subscribers per a terminal is low, and not to calculate the terminal or application CEI values, respectively, if the terminal could not be determined or if the given application has a small subscriber base.

8. A system (1) according to any one of the preceding claims; characterized by the processor (3) which is configured to perform the classification of measurements into “good”, “moderate” and “poor” categories through the location in which their sub-metrics are located within the user community differentiated by the relevant application, terminal, and technology; to determine the percentage of “poor” examples after the labelling process; and to mainly collect these percentages for the groupings in the form of sub-metric, user, technology, terminal, and application in order to obtain CEI values related to the groupings in the form of user, technology, terminal, and application.

9. A system (1) according to any one of the preceding claims; characterized by the processor (3) which is configured to model quality of experience by using customer feedback in the form of “complaints” and QoS oriented metric of CEI, and to divide customers or FWA subscribers into two groups as the customers who have not filed any complaints during a certain period of time and the others.

10. A system (1) according to any one of the preceding claims; characterized by the processor (3) which is configured to develop a transition-to-complaint model for a quantitative quality of experience in the form of a statistical QoE, threshold function, probability of complaint and complaint by using current subscriber complaint data.

11. A system (1) according to any one of the preceding claims; characterized by the processor (3) which is configured to assume that there is a probability of a user filing a complaint at any time depending on the hidden QoE values with the developed model.

12. A system (1) according to any one of the preceding claims; characterized by the processor (3) which is configured to develop a model by handling complaint data collectively and focusing on whether a complaint has been filed or not during a certain period of time.

13. A system (1) according to any one of the preceding claims; characterized by the processor (3) which is configured to provide an unknown threshold function in order to identify the relation between QoE and the probability of complaint.

14. A system (1) according to any one of the preceding claims; characterized by the processor (3) which is configured to determine the probability of a complaint to occur by the proportion of cases where QoE does not exceed a certain threshold.

15. A system (1) according to any one of the preceding claims; characterized by the processor (3) which is configured to predict the probability of complaint for a group of users by taking into account the proportion of data corresponding to users who have filed a complaint.

16. A system (1) according to any one of the preceding claims; characterized by the processor (3) which is configured to measure the probability of a user complaining for the first time at a certain time by the probability that the QoE stays between two time-dependent thresholds.

17. A system (1) according to any one of the preceding claims; characterized by the processor (3) which is configured to consider QoE not as time-varying but as a random variable that is measured in general.

18. A system (1) according to any one of the preceding claims; characterized by the processor (3) which is configured to obtain the QoE model from the CEI by utilizing KPIs.

19. A system (1) according to any one of the preceding claims; characterized by the processor (3) which is configured to measure the information in the form of time spent, total volume, round-trip time, efficiency, speed, and access time by using the Kolmogorov-Smirnov statistic between the relevant empirical Cumulative Distributions and the discrepancy between the complainants and the others.

20. A system (1) according to any one of the preceding claims; characterized by the processor (3) which is configured to adjust the available CEI values corresponding to the respective percentiles, depending on the proportion of complaint data in each group, and to match them for QoE; and to consider the original CEI as a basis for QoE in order to be aligned with the previously mentioned model.

21. A system (1) according to any one of the preceding claims; characterized by the processor (3) which is configured to perform the matching as adding an appropriate value to all available CEIs in one group and removing another appropriate value from all available CEIs in the other group, therefore to bring the corrected CEI values corresponding to the relevant percentiles to a common point and to enable this point to be decided by the CEI value in the percentile corresponding to the proportion of complaint data for all data.

22. A system (1) according to any one of the preceding claims; characterized by the processor (3) which is configured to develop a measurement into which theCEI is converted, called a-CEI, as a representation of QoE and to decide on CEI adjustments by taking into account the change in a performance measurement that is determined to end.

Images