JP2024023007A - Information processing device, information processing method and program - Google Patents

Abstract

To lighten the burden on an operator who is requested by a user.SOLUTION: The present invention relates to an information processing device that has: an acquisition part which acquires user data including telemetry data related to a terminal that a user uses and/or personal data related to the user; a reception part which receives input of answer data, showing details of an answer obtained by questioning a user who receives a procedure request, from an operator terminal; a storage part which stores a model so generated to output a score showing the degree of illegality of the procedure request that the user makes based upon the user data and answer data; an estimation part which applies the user data acquired by the acquisition part to the model to estimate a first score, and applies the user data acquired by the acquisition part and the answer data received by the reception part to the model to estimate a second score; and an output part which outputs the user data, first score and second score to the operator terminal.SELECTED DRAWING: Figure 5

Description

The present invention relates to an information processing device, an information processing method, and a program.

Currently, various services for terminal failures are being developed. As an example, a service is known in which the owner of a mobile phone terminal can pay a fixed amount every month to have the terminal repaired or replaced when the terminal malfunctions or is damaged. Note that, for example, Patent Document 1 is known as a technique that enables smooth repair acceptance when a problem occurs with a device.

Japanese Patent Application Publication No. 2005-135299

When an operator of a business providing the above-mentioned services receives a request for repair or replacement from a user, it is necessary to confirm that the request is a legitimate request and not for fraudulent purposes. However, performing such confirmation may be a burden to the operator.

Therefore, an object of the present invention is to provide a technique that can reduce the burden on the operator who receives requests from users.

An information processing device according to one aspect of the present invention includes an acquisition unit that acquires user data including telemetry data regarding a terminal used by a user and/or personal data regarding the user, and an acquisition unit that asks questions of the user who has received a procedure request. a reception unit that accepts input of response data indicating the content of the response obtained by the procedure from an operator terminal; A first score is estimated by applying the user data acquired by the acquisition unit to the model, and the user data acquired by the acquisition unit and the answer data received by the reception unit are It has an estimating section that estimates a second score by applying it to the model, and an output section that outputs the user data, the first score, and the second score to an operator terminal.

According to the present invention, it is possible to provide a technique that can reduce the burden on an operator who receives requests from users.

1 is a diagram showing a configuration example of a call center system according to an embodiment of the present invention. It is a diagram showing an example of the hardware configuration of a data acquisition device and a call center device. It is a diagram showing an example of the hardware configuration of an operator terminal. FIG. 2 is a diagram showing an example of a functional block configuration of a data acquisition device. It is a diagram showing an example of a functional block configuration of a call center device. It is a figure showing an example of functional block composition of an operator terminal. It is a flowchart which shows an example of the processing procedure performed by a call center apparatus. FIG. 3 is a diagram showing an example of an operator screen displayed on an operator terminal. It is a figure which shows the update example of a score.

Embodiments of the present invention will be described with reference to the accompanying drawings. In addition, in each figure, those with the same reference numerals have the same or similar configurations. In the following, an example in which the call center system is applied to a call center that receives inquiries regarding repair, replacement, etc. of mobile devices will be described, but the present embodiment is not limited thereto. In addition to mobile devices, the call center system according to the present embodiment is capable of handling various devices used by customers (also referred to as users), such as so-called IoT home appliances or smart home appliances such as speakers and vacuum cleaners that are connected to a communication network. It can be applied to call centers in general that provide support regarding equipment.

<System configuration>
FIG. 1 is a diagram showing a configuration example of a call center system according to an embodiment of the present invention. A call center system 1 according to the present embodiment is a system for receiving inquiries regarding terminals used by customers (hereinafter referred to as customer terminals 10) and providing support to the customers. The customer terminal 10 is connected to a data acquisition device 20 via a communication network N. The call center system 1 according to the present embodiment includes a call center device 30 that is connectable to the data acquisition device 20 via a communication network N, and an operator terminal 40 that is connectable to the call center device 30.

The communication network N is a communication network capable of mutually transmitting and receiving information between the data acquisition device 20 and the call center device 30, and between the call center device 30 and the operator terminal 40, and includes the Internet, LAN, private line, and telephone. It can be any line, corporate network, mobile communication network, Bluetooth (registered trademark), Wi-Fi (Wireless Fidelity), other communication lines, or a combination thereof, and whether it is wired or wireless. No questions asked.

The customer terminal 10 is a device with a communication function that is connected to the communication network N and used. Specifically, the customer terminal 10 may include an information processing device such as a smartphone, a tablet terminal, a personal computer (PC), or a notebook PC, or a communication device such as a mobile phone. Alternatively, the customer terminal 10 may include any device that is subject to repair/replacement services, such as a so-called IoT home appliance or smart home appliance such as a speaker, a vacuum cleaner, a cooking appliance, etc. that is connected to the communication network N and used.

The data acquisition device 20 is a server device that acquires telemetry data from the customer terminal 10 and stores it in association with identification information of the customer terminal 10. Telemetry data is data that indicates the status of the customer terminal 10, and specifically includes information regarding power supply (charge, discharge, etc.), information regarding communication status (radio wave status signal: within range, out of range, strength, etc.), Memory usage status, various setting value setting or change events, installed application startup and/or termination events, application operation details and/or operation history, application installation date and time, website access information, etc. The occurrence time and/or end time of the event, location information detected by GPS etc., detection data (for example, data on the detection of an impact, etc.) by a sensor (for example, an acceleration sensor) installed in the customer terminal 10, various hardware ( Examples include data such as the operating status of buttons, sensors, cameras, etc.) and chip information.

Here, in the customer terminal 10, it is possible to set whether or not to collect telemetry data and the telemetry data that can be collected. You may obtain it. The data acquisition device 20 may continuously acquire telemetry data from the customer terminal 10, or may acquire it in a snapshot manner at a predetermined period. In addition, the data acquisition device 20 stores only the telemetry data acquired within a predetermined period (for example, the most recent hour, several hours, one day, several days, etc.), and when new telemetry data is received, old Telemetry data may be deleted sequentially.

The call center device 30 is a device used in the call center system 1. The call center device 30 acquires telemetry data regarding the customer terminal 10 used by the customer who received the inquiry from the data acquisition device 20, and causes the operator terminal 40 to display the telemetry data. Further, the call center device 30 causes the operator terminal 40 to display various data (hereinafter referred to as "personal data") regarding the individual customer who received the inquiry. Personal data includes, for example, the customer's identification information (hereinafter referred to as "customer ID"), the customer's name, age, date of birth, address, the period of the communication service contract, and the name of the customer terminal 10 purchased. The information may include identification information, the date of purchase of the customer terminal 10, the store where the customer terminal 10 was purchased, the period of use of the customer terminal 10, the history of repairs or replacements of the customer terminal 10 in the past, and the like.

In addition, when receiving a procedure request for repair or replacement of the customer terminal 10 (hereinafter referred to as a "service usage application") from a customer, the call center device 30 transmits telemetry data acquired from the customer terminal 10 and the customer's personal information. Based on the data, a score indicating the degree of fraud in the service usage application is estimated, and the estimated score is displayed on the operator terminal 40. In addition to the telemetry data and the customer's personal data, the call center device 30 also uses information based on data indicating the content of answers obtained by the operator asking questions of the customer (hereinafter referred to as "response data"). The score is updated and the updated score is displayed on the operator terminal 40. The operator may ask the customer, for example, when, where, and how in response to a service application requesting that a replacement product be sent because the customer's terminal 10 is damaged. ) Questions include whether the customer terminal 10 is damaged.

The data acquisition device 20 and the call center device 30 may each be configured from one or more servers, or may be configured using a virtual server running on a hypervisor, or may be configured using a virtual server running on a hypervisor. It may be configured using a cloud server. The data acquisition device 20 and the call center device 30 may each be called an information processing device. Further, the data acquisition device 20 and the call center device 30 may be integrated.

The operator terminal 40 is a terminal device used by an operator who responds to inquiries from customers at a call center or the like. Note that a plurality of operator terminals 40 may be provided in the call center system 1.

<Hardware configuration>
FIG. 2 is a diagram showing an example of the hardware configuration of the data acquisition device 20 and the call center device 30. The data acquisition device 20 and the call center device 30 are configured by, for example, a computer with high processing power. The data acquisition device 20 and the call center device 30 include processors 21 and 31 such as a CPU (Central Processing Unit) and a GPU (Graphical Processing Unit), memory, and storage such as an HDD (Hard Disk Drive) and/or an SSD (Solid State Drive). It has devices 22, 32, communication interfaces (IF) 23, 33 for performing wired or wireless communication, input devices 24, 34 for accepting input operations, and display devices 25, 35 for displaying information. The input devices 24 and 34 are, for example, a keyboard, a touch panel, a mouse, or the like. The display devices 25 and 35 are, for example, liquid crystal displays or organic EL displays.

FIG. 3 is a diagram showing an example of the hardware configuration of the operator terminal 40. As shown in FIG. The operator terminal 40 is configured by, for example, a computer such as a personal computer (PC), a notebook PC, or a tablet terminal. The operator terminal 40 includes a processor 41 such as a CPU (Central Processing Unit) and a GPU (Graphical Processing Unit), a memory, a storage device 42 such as an HDD (Hard Disk Drive) and/or an SSD (Solid State Drive), and wired or wireless communication. It has a communication interface (IF) 43 for performing this, an input device 44 for accepting input operations, and a display device 45 for displaying information. The input device 44 is a keyboard, touch panel, mouse, or the like. The display device 45 is a liquid crystal display, an organic EL display, or the like. Further, the operator terminal 40 may further include a microphone 46 that accepts voice input, a speaker 47 that outputs voice, and a camera 48 that captures images.

<Functional block configuration>
FIG. 4 is a diagram showing an example of the functional block configuration of the data acquisition device 20. As shown in FIG. The data acquisition device 20 includes a storage section 200 that stores telemetry data 201, a communication section 211, a telemetry data management section 212, and a telemetry data extraction section 213.

Among these, the storage unit 200 can be realized using the storage device 22 included in the data acquisition device 20. The communication unit 211 can be realized using the communication interface 23 included in the data acquisition device 20. Further, the telemetry data management section 212 and the telemetry data extraction section 213 can be realized by the processor 21 of the data acquisition device 20 executing a program stored in a storage device. Note that the program can be stored in a storage medium. The storage medium storing the program may be a non-transitory computer readable medium. The non-temporary storage medium is not particularly limited, but may be a storage medium such as a USB memory or a CD-ROM.

The telemetry data management unit 212 manages the timing and period of receiving telemetry data from the customer terminal 10, and stores the received telemetry data in the storage unit 200 in association with identification information of the customer terminal 10 (hereinafter referred to as terminal ID). . As the terminal ID, for example, an IMEI (International Mobile Equipment Identifier) or an IP address can be used. Further, the telemetry data management unit 212 manages the storage period of telemetry data in the storage unit 200.

Telemetry data extraction section 213 extracts telemetry data regarding a specific customer terminal 10 in response to a request from call center device 30 and transmits it from communication section 211 to call center device 30 .

FIG. 5 is a diagram showing an example of the functional block configuration of the call center device 30. As shown in FIG. The call center device 30 includes a storage section 300, a communication section 311, an acquisition section 312, a reception section 313, an estimation section 314, and an output section 315. Further, the call center device 30 may further include an estimation model creation section 316.

The storage unit 300 can be realized using the storage device 32 included in the call center device 30. The communication unit 311 can be realized using the communication interface 33 included in the call center device 30. In addition, the acquisition unit 312 , the reception unit 313 , the estimation unit 314 , the output unit 315 , and the estimation model creation unit 316 are configured so that the processor 31 of the call center device 30 executes a program stored in the storage device 32 . This can be realized by The program can be stored in a storage medium. The storage medium storing the program may be a computer-readable non-transitory storage medium. The non-temporary storage medium is not particularly limited, but may be a storage medium such as a USB memory or a CD-ROM.

The storage unit 300 stores a customer information DB 301 and an estimation model 302.

The customer information DB 301 stores personal data of customers. Note that all or part of the information stored in the customer information DB 301 may be stored in another information processing device that can communicate with the call center device 30. The call center device 30 may acquire the customer's personal data by accessing the other information processing device as necessary.

The estimation model 302 is a mathematical model constructed to estimate a score indicating the degree of fraud in a service usage application. The estimation model 302 outputs a score when telemetry data and personal data are input. Further, the estimation model 302 outputs a score when telemetry data, personal data, and response data are input. In the following description, telemetry data and personal data will be collectively referred to as user data. That is, user data includes telemetry data and/or personal data.

Such an estimation model 302 includes past service usage application history (including answer data obtained from questions when accepting a service usage application and data indicating whether the service usage application was legitimate or fraudulent); It can be constructed by having a model learn patterns with user data. As for user data, a method of defining a correlation between telemetry data and a service usage application using a function or the like can be cited as a mathematical model. Alternatively, the estimation model 302 may be constructed using commonly used machine learning such as decision tree-based gradient boosting, Factorization-Machine, and deep learning.

The acquisition unit 312 acquires user data including telemetry data regarding the customer terminal 10 used by the customer and/or personal data regarding the customer. The acquisition unit 312 may acquire telemetry data from the data acquisition device 20 and store the acquired telemetry data in the storage unit 300. Further, the acquisition unit 312 may acquire personal data from the customer information DB 301.

Further, the acquisition unit 312 may acquire telemetry data acquired within a predetermined period (for example, the most recent hour, several hours, one day, several days, etc.) from the data acquisition device 20.

The reception unit 313 receives various inputs from the operator terminal 40. For example, the reception unit 313 receives an input from a customer indicating that a service usage application has been received, and inputs such as the customer's name. The reception unit 313 also receives input from the operator terminal 40 of response data indicating the content of the response obtained by asking a question to a customer who has applied for service use. Further, the receiving unit 313 may update the answer data according to the answer obtained by asking the customer an additional question. The response data may include a series of responses obtained from the customer. Further, the reception unit 313 may receive input of answer data (that is, automatically perform the input) by performing voice recognition on voice data of an answer obtained by asking a question to a customer. It becomes possible for the operator to save the effort of inputting response data into the operator terminal 40.

The estimation unit 314 applies the user data or the user data and response data to the estimation model 302 to estimate a score indicating the degree of fraud in the service usage application. Specifically, the estimation unit 314 extracts or calculates predetermined feature amounts from the user data (or user data and response data), and inputs these feature amounts to the prediction model 304 to obtain the score estimation result. .

The feature quantities input to the estimation model 302 include the telemetry data itself, the occurrence frequency and occurrence state of various events in the telemetry data (communication state, memory state, etc.), and event strings (multiple data showing the occurrence frequency and occurrence state of events). ) by methods used in natural language processing such as TF-IDF, numerical values obtained by converting character strings included in telemetry data and response data based on predetermined definition information (for example, numerical values indicating locations, etc.), etc. can be used.

Further, by applying the user data acquired by the acquisition unit 312 to the estimation model 302, a score (first score) indicating the degree of fraud in the service usage application is estimated. Furthermore, the estimation unit 314 estimates a score (second score) indicating the degree of fraud in the service usage application by applying the user data and the response data received by the reception unit 313 to the estimation model 302.

The score (first score) estimated by applying user data to the estimation model 302 is estimated using user data obtained after receiving a service usage application from a customer and before asking the customer a question. It may be the score that is given. Furthermore, the score (second score) estimated by applying user data and response data to the estimation model 302 is obtained by the user after receiving a service usage application from the customer and before asking the customer a question. The score may be estimated using data and answer data obtained by asking questions to the customer.

Further, the estimation unit 314 may update the score (second score) by applying the user data acquired by the acquisition unit 312 and the updated answer data to the estimation model 302.

Furthermore, the estimating unit 314 may estimate (output) the degree of contribution of each piece of user data used for estimation with respect to the estimated score (first score). Furthermore, the estimating unit 314 may estimate the degree of contribution of each of the user data and response data used for estimation, regarding the estimated score (second score).

Contribution degree refers to the degree to which each feature input to the estimation model 302 (i.e., each user data and each response data) influences the output value (i.e., score) from the estimation model 302. The larger the value, the greater the degree of contribution. For example, when feature quantities 1 to 10 are input to the estimation model 302, the prediction accuracy of the estimation model 302 hardly changes even if feature quantity 1 is changed, but the prediction accuracy of the estimation model 302 changes when feature quantity 2 is changed. In the case where there is a large change in the feature quantity 2, it can be said that the contribution degree of the feature quantity 2 is greater than that of the feature quantity 1. The degree of contribution for each feature can be obtained, for example, by using an existing library corresponding to the estimation model 302. The degree of contribution may also be called the degree of importance.

Furthermore, the estimating unit 314 may estimate (output) the direction of contribution of each piece of user data used for estimation with respect to the estimated score (first score). Furthermore, the estimating unit 314 may estimate the direction of contribution of each of the user data and answer data used for estimation with respect to the estimated score (second score).

The contribution direction is a direction in which each feature input to the estimation model 302 (i.e., each user data and each response data) is in a negative direction (in a direction in which the output value is decreased) with respect to the output value (i.e., score) of the estimation model 302. This is an index indicating whether the output value is contributing to the output value or in the positive direction (direction of increasing the output value). For example, when the output value when the features 1 to 3 are input to the estimation model 302 is 60, the contribution of feature 1 is 50, the contribution of feature 2 is 30, and the contribution of feature 3 is 60. Assume that the contribution of is -20. In this case, it can be seen that feature quantities 1 and 2 contribute in the positive direction, and feature quantity 3 contributes in the negative direction. The contribution direction can be obtained, for example, by using an existing library corresponding to the estimation model 302.

The output unit 315 outputs the user data and the score (first score and/or second score) estimated by the estimation unit 314 to the operator terminal 40. Further, the output unit 315 may output the score updated by the estimation unit 314 (the updated second score). In addition, the output unit 315 outputs the degree of contribution of each piece of user data regarding the score (first score), and the user data and response data regarding the score (second score, updated second score) output from the estimation unit 314. The degree of contribution of each of these may be output.

In addition, the output unit 315 outputs information on the contribution direction of each user data regarding the score (first score) and the contribution direction of each user data and answer data regarding the score (second score, updated second score). The contribution direction may be output in a distinguishable manner depending on whether the contribution direction is positive or negative. For example, the output unit 315 may output display data such that each user data or each user data and response data is displayed in different colors depending on whether the contribution direction is positive or negative. good.

The estimated model creation unit 316 constructs or reconstructs the estimated model 302 using user data and response data. Specifically, the estimation model creation unit 316 calculates the user data based on the user data and response data, and the final judgment result as to whether the customer's service application was a legitimate request or an unauthorized request. and response data and score patterns are created as learning data, and the estimation model 302 is made to learn. The estimated model creation unit 316 stores the estimated model 302 constructed or reconstructed through learning in the storage unit 300.

FIG. 6 is a diagram showing an example of the functional block configuration of the operator terminal 40. The operator terminal 40 includes a storage section 400, a communication section 411, a display control section 412, and an input reception section 413.

The storage unit 400 can be realized using the storage device 42 included in the operator terminal 40. The communication unit 411 can be realized using the communication interface 43 included in the operator terminal 40. Further, the display control section 412, the input reception section 413, and the report creation section 414 can be realized by the processor 41 of the operator terminal 40 executing a program stored in the storage device 42. The program can be stored in a storage medium. The storage medium storing the program may be a computer-readable non-transitory storage medium. The non-temporary storage medium is not particularly limited, but may be a storage medium such as a USB memory or a CD-ROM.

The display control unit 412 controls the screen display on the display device 45. Specifically, the display control unit 412 displays a customer management screen on the display device 45 for allowing the operator to input information regarding a customer applying for service use, and for presenting estimated scores and user data to the operator. let The customer management screen may be configured on the operator terminal 40 side based on a program stored in the storage device 42, or may be displayed using a web browser based on data sent from the call center device 30. It may be something that is done.

The input receiving unit 413 receives a signal corresponding to an operation on the input device 44, and executes an operation on the customer management screen or displays information on the screen in response to the signal.

<Processing procedure>
FIG. 7 is a flowchart illustrating an example of a processing procedure performed by the call center device 30.

In step S10, the call center operator inputs into the operator terminal 40 the fact that a service application has been received from the customer and the details of the service application (for example, failure, damage, loss, submergence, theft, etc.). The reception unit 313 of the call center device 30 receives, via the operator terminal 40, the fact that a customer has applied for service use and the contents of the service use application.

In step S20, the acquisition unit 312 of the call center device 30 accesses the data acquisition device 20 and acquires telemetry data for a predetermined period at the customer terminal 10 used by the customer who has applied for service. The predetermined period is not particularly limited, but may be, for example, the past day, the past week, etc. Further, the predetermined period may differ depending on the item of telemetry data. For example, the acquisition unit 312 may acquire position information for the past month, and acquire data that does not change, such as the model name and storage capacity, for the most recent hour or the latest data. Further, the acquisition unit 312 accesses the customer information DB 301 and acquires personal data corresponding to the customer ID of the customer.

In step S30, the estimation unit 314 of the call center device 30 estimates a score by applying the user data (telemetry data and/or personal data) acquired in step S20 to the estimation model 302. This score may be called an initial score or a primary score because it is a score before asking questions to the customer who has applied for the service. In the following explanation, this will be referred to as a "primary score." Furthermore, the estimation unit 314 outputs the degree of contribution and direction of contribution of each piece of user data used for estimation, regarding the estimated primary score. The degree of contribution and direction of contribution in the user data used for estimation can be output by using, for example, a library called SHAP (SHApley Additive exPlanations), but is not limited thereto.

Further, the output unit 315 outputs display data for displaying the user data and the primary score estimated by the estimation unit 314 on the operator terminal 40. The display data is transmitted to the operator terminal 40 by the communication unit 311.

In step S40, the call center operator asks the customer questions regarding the accident situation and the like, and inputs the answers to the questions into the operator terminal 40. The reception unit 313 of the call center device 30 receives response data indicating the content of the customer's response to the question via the operator terminal 40. The content that the operator should ask the customer may be displayed on the operator terminal 40. Further, different question contents may be displayed on the operator terminal 40 depending on the contents of the service usage application. The output unit 315 may output display data indicating the content of the question that the operator should ask the customer, depending on the content of the service usage application received in the processing procedure of step S10.

In step S50, the estimation unit 314 of the call center device 30 estimates a score by applying the user data acquired in step S20 and the response data received in step S40 to the estimation model 302. Since the score is a score after asking questions to the customer who has applied for the service, it may be called an updated score, a secondary score, or the like. In the following explanation, this will be referred to as a "secondary score." Furthermore, the estimation unit 314 outputs the degree of contribution and direction of contribution of each piece of user data used for estimation, regarding the estimated secondary score.

Further, the output unit 315 outputs display data for displaying the user data and the secondary score estimated by the estimation unit 314 on the operator terminal 40. The display data is transmitted to the operator terminal 40 by the communication unit 311.

In step S60, if the call center device 30 does not end the process, the call center device 30 returns to step S40 and repeats the processing procedures of step S40 and step S50. The operator asks the customer additional questions, and the call center device 30 updates the secondary score using the answer data obtained from the additional questions. Further, when the call center device 30 receives an instruction via the operator terminal 40 that the customer service has been completed, the call center device 30 ends the process.

<Example of display on operator terminal>
FIG. 8 is a diagram showing an example of an operator screen displayed on the operator terminal 40. In the display area 100, acquired user data (customer's personal data and telemetry data acquired from the customer terminal 10) is displayed side by side for each item. The display area 101 displays the estimated score, the degree of contribution and the direction of contribution for each user data regarding the score. Display area 102 displays an input column for inputting the content of the answer obtained by asking the customer a question. The input field may allow input using a pull-down menu.

FIG. 9 is a diagram showing an example of updating scores. It is assumed that the score means that a higher value means a higher probability of fraud. First, the estimation unit 314 of the call center device 30 estimates the primary score based on the user data acquired when accepting the application for service use, and estimates the degree of contribution and direction of contribution for each user data regarding the estimated primary score. do.

In the example of FIG. 9, the estimated primary score is 0.6, and the user data that contributes positively to the score are ordered by the storage capacity of the customer terminal 10 and the service contract. period. Furthermore, the user data that contribute negatively to the score are the past service usage application history and the customer's age, in descending order of the degree of contribution in the negative direction.

Subsequently, the operator asks the customer questions and inputs the customer's answers into the operator terminal 40. The estimation unit 314 of the call center device 30 estimates the secondary score by applying the user data and response data to the estimation model 302. Furthermore, the estimating unit 314 estimates the contribution degree and contribution direction for each user data regarding the estimated secondary score.

In the example of FIG. 9, the estimated secondary score is 0.7, and the user data that contributes positively to the score are the date and time of damage, the memory of the customer terminal 10, and the user data that contributes positively to the score. capacity and service contract period. Furthermore, the user data that contribute negatively to the score are the past service usage application history and the customer's age, in descending order of the degree of contribution in the negative direction. As shown in FIG. 9, the score changes from 0.6 to 0.7 depending on the customer's answers to the questions.

Subsequently, the operator asks the customer further questions and inputs the customer's answers into the operator terminal 40. The estimation unit 314 of the call center device 30 updates the secondary score by applying the user data and response data to the estimation model 302. Furthermore, the estimating unit 314 estimates the contribution degree and contribution direction for each user data regarding the updated secondary score.

As shown in FIG. 9, the score changes from 0.6 to 0.7 depending on the answer to hearing step 1. This is because the information regarding hearing step 1 included in the telemetry data was different from the customer's answer. Similarly, depending on the answer to hearing step 2, the score changes from 0.7 to 0.8. This is because the information regarding hearing step 2 included in the telemetry data was different from the customer's answer. In this way, if there is doubt that the customer is telling the truth, the score will change in a positive direction.

<Summary>
According to the embodiment described above, when an application for service use is received from a customer, the operator terminal 40 displays a score indicating the degree of fraud in the application for service use. This allows the operator to objectively judge whether the application for service use is legitimate or is likely to be fraudulent while looking at the screen. This makes it possible to reduce the burden on the operator who receives requests from customers. Further, in this embodiment, the score is updated by the operator inputting the contents of the answers from the customers into the operator terminal 40. This allows the operator to more accurately recognize whether the customer's application for service is legitimate or is likely to be fraudulent.

The embodiments described above are intended to facilitate understanding of the present invention, and are not intended to be interpreted as limiting the present invention. The flowcharts, sequences, and elements included in the embodiments, as well as their arrangement, materials, conditions, shapes, sizes, etc., described in the embodiments are not limited to those illustrated and can be changed as appropriate. Further, it is possible to partially replace or combine the structures shown in different embodiments.

1... Call center system, 10... Customer terminal, 20... Data acquisition device, 21, 31, 41... Processor, 22, 32, 42... Storage device, 23, 33, 43... Communication interface (IF), 24, 34, 44 ...Input device, 25, 35, 45... Display device, 30... Call center equipment, 40... Operator terminal, 46... Microphone, 47... Speaker, 48... Camera, 200, 300, 400... Storage section, 201... Telemetry data, 211・311 411...Communication department, 212...Telemetry data management section, 213...Telemetry data extraction section, 301...Customer information database (DB), 302...Estimation model, 312...Acquisition section, 313...Reception section, 314...Estimation section , 315... Output section, 316... Estimation model creation section, 412... Display control section, 413... Input reception section

Claims (9)

an acquisition unit that acquires user data including telemetry data regarding a terminal used by a user and/or personal data regarding the user;
a reception unit that receives input from an operator terminal of response data indicating the content of the response obtained by asking a question to the user who has received the procedure request;
a storage unit that stores a model generated to output a score indicating the degree of fraud of the procedure request made by the user based on the user data and the response data;
A first score is estimated by applying the user data acquired by the acquisition unit to the model, and the user data acquired by the acquisition unit and the answer data received by the reception unit are applied to the model. an estimation unit that estimates the second score by applying the
an output unit that outputs the user data, the first score, and the second score to the operator terminal;
An information processing device having: The first score is estimated using the user data obtained after receiving a procedure request from the user and before asking the user a question,
The second score includes the user data obtained after receiving a procedure request from the user and before asking the user a question, and the answer data obtained by asking the user a question. estimated using
The information processing device according to claim 1. The reception unit updates the answer data according to the answer obtained by asking the user an additional question,
The estimation unit updates the second score by applying the user data acquired by the acquisition unit and the updated answer data to the model,
the output unit outputs the updated second score;
The information processing device according to claim 1. The estimation unit estimates the degree of contribution of each of the user data regarding the first score, and the degree of contribution of each of the user data and the answer data regarding the second score,
The output unit outputs a degree of contribution of each of the user data regarding the first score and a degree of contribution of each of the user data and the answer data regarding the second score.
The information processing device according to claim 1. The estimation unit estimates the contribution direction of each of the user data regarding the first score, and the contribution direction of each of the user data and the answer data regarding the second score,
The output unit is configured to determine the contribution direction of each of the user data regarding the first score and the contribution direction of each of the user data and the answer data regarding the second score, when the contribution direction is positive and when the contribution direction is negative. Output in an identifiable manner with
The information processing device according to claim 1. The reception unit receives input of the answer data by performing voice recognition on voice data obtained by asking the user a question.
The information processing device according to claim 1. The information processing device according to claim 1, wherein the acquisition unit acquires telemetry data acquired within a predetermined period from a data acquisition device. obtaining user data including telemetry data regarding a terminal used by a user and/or personal data regarding said user;
a step of accepting from an operator terminal input of answer data indicating the content of the answer obtained by asking a question to the user who has received the procedure request;
storing in a storage unit a model generated to output a score indicating the degree of fraud of the procedure request made by the user based on the user data and the response data;
A first score is estimated by applying the acquired user data to the model, and a second score is estimated by applying the previously acquired user data and the received response data to the model. the step of
outputting the user data, the first score, and the second score to the operator terminal;
An information processing method performed by an information processing device, including: obtaining user data including telemetry data regarding a terminal used by a user and/or personal data regarding said user;
a step of accepting from an operator terminal input of answer data indicating the content of the answer obtained by asking a question to the user who has received the procedure request;
storing in a storage unit a model generated to output a score indicating the degree of fraud of the procedure request made by the user based on the user data and the response data;
A first score is estimated by applying the acquired user data to the model, and a second score is estimated by applying the acquired user data and the received response data to the model. step and
outputting the user data, the first score, and the second score to the operator terminal;
A program that causes a computer to execute

Images