JP2014160320A - Customer confirmation support system, customer confirmation support method, and customer confirmation support program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a customer confirmation support system, a customer confirmation support method, and a customer confirmation support program for supporting provision of an appropriate service by confirming a customer state.SOLUTION: A control part 21 of a customer confirmation support server 20 performs pattern matching for an obtained transaction request to output the matching result, in which an item that is in match is discriminated from an item that is not in match, in the pattern matching, so as to call attention. In a case where it is determined that a countermeasure is required on a customer reception terminal 11 or operator terminal 15, the control part 21 executes hearing processing, in which information related to a question for confirming a customer itself is output in accordance with the matching result. If determined that there is a problem, the control part 21 suspend a transaction, or executes transaction processing if determined that there is no problem. If determined, after the transaction, that there is no problem, the control part 21 updates an individual pattern, or, if determined then that there is a problem, updates a call-attention pattern.

Description

  The present invention relates to a customer confirmation support system, a customer confirmation support method, and a customer confirmation support program for supporting confirmation of a customer state when providing a service.

  Transfer fraud that induces illegal transfers is a social problem. In the transfer fraud, the victim who misrecognizes the remittance destination due to fictitious facts is requested to transfer to the account specified by the fraudster himself. Then, the technique for suppressing such transfer is examined (for example, refer patent documents 1-4).

  The technique described in Patent Document 1 includes a face detection unit that detects the presence or absence of a facial image that allows the user to determine the human phase, acquires a withdrawal account input by an automatic transaction apparatus, and the face detection unit detects Without the withdrawal account obtaining means, the transaction of the automatic transaction apparatus is stopped.

  In the technique described in Patent Document 2, the user telephone terminal records call voice data and transmits it to the call content determination server. The call content determination server collates the voice characteristics based on the call voice data received from the user telephone terminal and the voice data of the pre-registered fraudster, converts the call voice data to text, and is included in the text-converted call contents The account information to be checked is matched with the fraudulent account information registered in advance. Then, the matching rate between the keyword registered in advance and the keyword included in the text-converted call content is calculated, and the determination result information whose determination level is determined based on the voice characteristics, the account information, and the keyword matching rate is used as the user telephone terminal. And a mail including the determination result information is generated and transmitted to a pre-registered notification destination.

  In addition, in the technique described in Patent Document 3, there is a possibility that the transfer destination account is a deposit account used for the transfer fraud based on the transaction history of the transfer destination account included in the transfer destination information. Judge whether the price is high. When it is determined that there is a high possibility that the bank account is a deposit account used for bank transfer fraud, a bank fraud alert message is displayed.

  Moreover, in the technique described in Patent Document 4, the host computer sets a plurality of transaction patterns based on dates within a month of a plurality of transactions stored in the transaction history database, and sets the set transaction patterns for each of a plurality of transaction information. Register with. Then, based on the account number of the user and the current transaction date when the transaction is performed, the transfer destination account number of the transaction information matching the transaction pattern is extracted from the transaction history database. Further, an average amount, an approximate amount, or a maximum amount of the transfer amount is calculated from a plurality of transaction information corresponding to the extracted transfer destination account number, and displayed on the display unit of these automatic teller machines. Then, the transaction details with high possibility of fraudulent user information and fraud are scored, and if this score exceeds a predetermined score, the user is warned.

JP 2010-79753 A (first page, FIG. 1) Japanese Patent Laid-Open No. 2008-210085 (first page, FIG. 1) JP 2007-272410 A (first page, FIG. 1) JP 2012-27615 A (first page, FIG. 1)

  In the technology described in the above patent document, a transaction that may cause a problem is detected by pattern matching. However, since the transaction status varies depending on the customer, appropriate detection may be difficult depending on the general pattern. Moreover, a transaction form may differ when there is no transaction record with a customer or depending on a customer's attribute. It is difficult to give a comprehensive alert to such diverse customer transactions. In addition, when confirming the customer status including the identity and appropriateness of transaction contents, the confirmation method may be lost.

  The present invention has been made in order to solve the above-mentioned problems, and its purpose is to confirm a customer state and support a customer confirmation support system, a customer confirmation support method, and a customer confirmation for supporting appropriate service provision. To provide support programs.

  (1) In the customer confirmation support system, the customer confirmation support method, and the customer confirmation support program that solve the above-mentioned problems, a unique pattern for each customer generated based on the customer service usage record is recorded in association with the customer identifier. Corresponding to the comparison result of the pattern comparison and the general information storage unit, the general information storage unit in which the general pattern generated based on the service usage record of the customer matching the customer attribute is recorded in association with the customer attribute A confirmation information storage unit for recording confirmation information and a control unit connected to a service providing channel, wherein the control unit acquires a customer identifier and request information of a customer in the service providing channel, and When a unique pattern is registered in the information storage unit in association with the customer identifier, the request information and the unique pattern are Turn comparison is performed, a comparison result is generated, and when the unique pattern is not registered in association with the customer identifier, the customer attribute of the customer is specified, and is associated with the customer attribute in the general information storage unit The general pattern is compared with the request information, a comparison result is generated, and confirmation information corresponding to the comparison result is extracted and output from the confirmation information storage unit. Thereby, when there is a transaction performance and it is desirable to call attention according to the feature at the time of service use for each customer, it is possible to call attention using a unique pattern for each customer. On the other hand, when there is no transaction record, or when there is a small transaction record and there is no feature when using the service for each customer, it is possible to call attention using a general pattern according to the customer attribute of the customer. As the unique pattern, a customer transaction pattern, or an individual pattern that represents a customer's identity or a characteristic amount of behavior can be used.

  (2) With respect to the customer confirmation support system, the unique pattern includes customer feature amount information, the feature information on the customer pattern is acquired in the service providing channel, and the feature amount information of the unique pattern And the feature amount information acquired in the service providing channel are preferably compared to generate a comparison result for the customer situation. As a result, the customer situation can be grasped by the feature amount such as the image and the sound, and the attention can be given.

  (3) The customer confirmation support system further includes a customer information storage unit in which related party information is recorded in association with a customer identifier, and the feature amount information of the unique pattern and the feature amount acquired in the service providing channel When it is determined that the customer is not the customer based on the comparison result with the information, the customer information storage unit obtains the customer party information, and the confirmation information storage unit confirms the information based on the party information. Is preferably extracted and output. As a result, it is possible to confirm the situation of the person's agent and call out an appropriate attention.

  (4) In the customer confirmation support system, the feature amount information in the unique pattern recorded in the unique information storage unit is registered for each service type, and the service type corresponding to the request information in the unique pattern is registered. It is preferable to identify feature quantity information and compare the identified feature quantity information with the feature quantity information acquired in the service providing channel. Thereby, the pattern matching can be performed in consideration of the situation in which the feature amount is acquired.

  (5) In the customer confirmation support system, when the information indicating that there is a problem in the service use record is not acquired, it is preferable to update the unique pattern based on the service use record. As a result, it is possible to maintain an appropriate unique pattern in response to changes in the customer situation.

  (6) The customer confirmation support system further includes a caution information storage unit in which a caution pattern for service provision is recorded, and when the information indicating that there is a problem in the service use result is acquired, the service use It is preferable to update the caution pattern based on the results. Thereby, it is possible to maintain an appropriate attention pattern in consideration of a problem occurrence state.

  (7) With respect to the customer confirmation support system, when request information for transfer to a transfer destination is acquired, use status information about the transfer destination is acquired and recorded in the use status information and the attention information storage unit It is preferable that the confirmation information corresponding to the comparison result of the pattern comparison is extracted and output from the confirmation information storage unit. Thereby, alerting can be performed in consideration of the situation of the bank account.

  (8) It is preferable that the customer confirmation support system is guided to a channel different from the service providing channel by the confirmation information. Thereby, even in a situation where the service providing channel cannot be confirmed accurately, appropriate alerting can be performed through another service providing channel.

  (9) About the said customer confirmation support system, it is preferable to change the weight of the element of a pattern comparison according to the request information from the said customer. Thereby, accurate alerting can be performed based on the request information.

  According to the present invention, it is possible to provide a customer confirmation support system, a customer confirmation support method, and a customer confirmation support program for confirming a customer state and supporting appropriate service provision.

The system schematic of one Embodiment of this invention. It is explanatory drawing of the data recorded on each information storage part of one Embodiment of this invention, (a) is an individual pattern information storage part, (b) is a customer pattern information storage part, (c) is general pattern information. Storage part, (d) is an attention pattern information storage part, (e) is explanatory drawing of the data recorded on the transaction request storage part. Explanatory drawing of the process sequence of one Embodiment of this invention. Explanatory drawing of the process sequence of one Embodiment of this invention. Explanatory drawing of the process sequence of one Embodiment of this invention. Explanatory drawing of the process sequence of one Embodiment of this invention. Explanatory drawing of the process sequence of one Embodiment of this invention. Explanatory drawing of the process sequence of one Embodiment of this invention. Explanatory drawing of the process sequence of one Embodiment of this invention.

  Hereinafter, an embodiment embodying the customer confirmation support method will be described with reference to FIGS. In the present embodiment, it is assumed that the state of a customer (transactioner) who conducts a transaction in a financial institution is confirmed and appropriate service provision (transaction) is supported. Here, there are cases where a customer visits a financial institution store and uses a window or an automated teller machine 12 or uses online transactions via the Internet.

  As shown in FIG. 1, in this embodiment, a customer confirmation support server 20 and a host system 30 connected to a transaction terminal 10 as a transaction channel (service provision channel) and an operator terminal 15 via a network are used.

  The host system 30 is a computer system that executes transactions in financial institutions. The host system 30 includes a customer information storage unit 32 and an account information storage unit 33 in order to manage customers and transactions of financial institutions.

  The customer information storage unit 32 stores information about customers of this financial institution. For example, data relating to attribute information such as name, address, age, and occupation is recorded for the customer code.

  In the account information storage unit 33, account information about an account opened in this financial institution is recorded. For example, for an account code, data relating to the customer code of the customer who holds the account, the account opening time, the presence or absence of an accident report, the balance, and the transaction history are recorded.

In the present embodiment, a window terminal 11, an automatic teller machine 12, and a user terminal 13 are used as a transaction terminal 10 that performs transactions with customers.
The window terminal 11 is a computer terminal installed at a window of a financial institution and used by a person in charge of the financial institution. The window terminal 11 includes a display unit such as a display and an input unit such as a keyboard and a pointing device. The window terminal 11 communicates with the host system 30 to request the host system 30 for a transaction with a customer. Further, in the store where the window terminal 11 is installed, an in-store camera for photographing customers visiting the store is provided. Then, the in-store camera shoots a customer at the window where the window terminal 11 is installed, and generates a customer image.

  The automatic teller machine 12 is a terminal that is installed in a store of a financial institution and performs transactions with customers through communication with the host system 30. The automatic teller machine 12 includes a card reader for specifying an account code from the inserted cash card. Furthermore, the automatic teller machine 12 includes a touch panel (an input unit and an output unit) for displaying various information to a customer who visits the store or making various inputs by the customer. Further, the automatic teller machine 12 includes a receiver for performing a conversation with a person in charge of a financial institution and a camera for photographing a user of the automatic teller machine 12.

  The user terminal 13 is a customer's computer terminal that performs online transactions using the Internet or the like. The user terminal 13 includes a display unit such as a display and an input unit such as a keyboard and a pointing device. Then, the user terminal 13 requests the host system 30 for a transaction by communicating with the host system 30 via the Internet.

  The operator terminal 15 is a computer terminal used by a person in charge of a financial institution that monitors transactions in the automatic teller machine 12 and the user terminal 13. The operator terminal 15 includes a display unit such as a display, an input unit such as a keyboard and a pointing device, a receiver for carrying out a conversation with a customer, and the like.

  Some of the transaction terminals 10 include a thermography device. This thermography device takes a temperature image (therm image) and measures the body temperature of a subject (customer). Some of the transaction terminals 10 include a microphone that collects customer's voice.

  The customer confirmation support server 20 is a computer system for confirming a customer situation and supporting an appropriate transaction. The customer confirmation support server 20 includes a control unit 21, an individual pattern information storage unit 22, a customer pattern information storage unit 23, a general pattern information storage unit 24, a caution pattern information storage unit 25, and a transaction request storage unit 26. In the present embodiment, the individual pattern information storage unit 22 and the customer pattern information storage unit 23 function as a unique information storage unit. Further, the customer pattern information storage unit 23 functions as a customer information storage unit, the general pattern information storage unit 24 functions as a general information storage unit, and the caution pattern information storage unit 25 functions as a caution information storage unit and a confirmation information storage unit.

  The control unit 21 includes a control unit such as a CPU (not shown) and a memory such as a RAM and a ROM, and supports the customer situation hearing according to the customer situation. By executing the customer confirmation support program for this purpose, the control unit 21 causes the request information acquisition unit 211, the confirmation management unit 212, the general collation unit 213, the individual collation unit 214, and the state collation unit 215, as shown in FIG. , Function as a transfer destination verification unit 216.

The request information acquisition unit 211 executes processing for acquiring a customer transaction request transmitted from the various transaction terminals 10.
The confirmation management unit 212 executes a process for managing the required hearing according to the customer.
The general verification unit 213 performs pattern verification with a general transaction pattern (general pattern) for a transaction request from a customer having no transaction history.

The individual verification unit 214 performs pattern verification on a transaction request from a customer with a transaction history with a transaction pattern (unique pattern) unique to the customer.
The state collation unit 215 performs pattern collation with a customer pattern (unique pattern) for a customer state having a transaction history.
The transfer destination verification unit 216 performs pattern verification on the status of the account at the transfer destination.

  As shown in FIG. 2A, the individual pattern information storage unit 22 records individual pattern data 220 relating to a normal transaction pattern (individual pattern) performed using each account established in a financial institution. The individual pattern data 220 is recorded when a transaction is performed using this account and a statistical value of each item for the transaction is calculated. In the individual pattern data 220, data relating to an account code, a transaction type, and a transaction pattern is recorded.

In the account code data area, data relating to an identifier (financial institution code, main branch code, deposit type, account number) for specifying an account opened in the financial institution is recorded.
In the transaction type data area, data relating to an identifier for specifying the type of transaction (for example, transfer or cash withdrawal) performed in a financial institution is recorded.

  In the transaction pattern data area, data relating to individual patterns for specifying the contents of transactions performed on a daily basis using this account is recorded. In this transaction pattern, a pattern relating to a transaction channel, a transaction place, a transaction time zone, a transaction amount, a customer, and the like is recorded.

  In the transaction channel, information for specifying a channel (the window terminal 11, the automatic teller machine 12, and the user terminal 13) used for the transaction using this account is recorded. In the present embodiment, information regarding the usage frequency is recorded for each transaction channel. Based on this usage frequency, it is possible to specify a transaction channel to be mainly used as a representative value, or specify a distribution status of a transaction channel as a distribution value.

  In the transaction place, information for specifying a place where the transaction is performed is recorded, such as a store having a window used for transactions using this account or a store where the automatic teller machine 12 is installed. In this embodiment, information regarding the usage frequency is recorded for each place. Based on this use frequency, it is possible to specify a place to be mainly used as a representative value, or to specify a distribution status of use places as a dispersion value.

  In the transaction time zone, information for specifying the time zone (every weekend, the end of every month, morning, afternoon, evening, etc.) in which transactions are performed using this account and the distribution status are recorded. In the present embodiment, information regarding the usage frequency is recorded for each time slot. Based on this use frequency, it is possible to specify a time zone that is mainly used as a representative value, or to specify a dispersion state of use time zones as a variance value.

  In the transaction amount, information for specifying an amount range (maximum amount or average amount) in a normal transaction using this account is recorded. In the present embodiment, information on the average amount (representative value) and the transaction amount distribution state (dispersion value) is recorded.

  In the business partner, an account code for specifying a transfer destination account when performing transfer using this account is recorded. In the present embodiment, information on the usage frequency is recorded for each business partner. Based on this usage frequency, it is possible to specify the main transfer destination as a representative value, or specify the distribution status of the transfer destination as a distribution value.

  As shown in FIG. 2B, the customer pattern information storage unit 23 records customer management data 230 relating to the situation of customers who have transactions with financial institutions. The customer management data 230 is recorded when a transaction is performed with the customer, or when an application or a hearing result is acquired from the customer. In the customer management data 230, data relating to a customer code, a customer pattern, a family situation, and an event situation is recorded.

In the customer code data area, data relating to an identifier for identifying a customer of this financial institution is recorded.
In the customer pattern data area, data relating to a unique pattern for confirming the identity of the customer and the state of the customer is recorded. In this embodiment, data relating to biometric authentication, gender, age, and personal status is recorded as a customer pattern.

In the biometric data area, data related to the feature amount used for the biometric authentication of the customer is recorded.
In the sex data area and the age data area, data relating to the sex and age of the customer are recorded, respectively.
In the personal situation data area, data relating to the customer's personal situation (body temperature, voice, facial expression, etc.) is recorded. Here, information related to body temperature and behavior (facial expressions calculated by face image processing, body movement feature quantities) and utterances in association with the transaction type (transfer, cash withdrawal, etc.) at the time of shooting as the service type Information (audio feature value) is recorded.

In the family situation data area, data related to the customer related information (family structure in this embodiment) is recorded. In this data area, data on family names, sexes, and ages are recorded.
In the event status data area, data related to the time and type of the customer's life event (marriage, moving, etc.) is recorded.

  As shown in FIG. 2C, the general pattern information storage unit 24 records general management data 240 related to transaction patterns (general patterns) that are generally performed in various transactions at financial institutions. Here, the general pattern is a pattern generated on the basis of an empirical rule or the like derived from a customer transaction record (service use record) that matches the customer attribute in association with the customer attribute. The general management data 240 is periodically updated when the transaction type, region, age group, and customer group are classified and the statistical value of the transaction method and transaction content is calculated. In the general management data 240, data related to transaction type, region, age group, customer group, and transaction pattern is recorded.

In the transaction type data area, data relating to an identifier for specifying the type of transaction (for example, transfer or cash withdrawal) performed in a financial institution is recorded.
In the region data area, data relating to an identifier for specifying the region to which the customer's address belongs is recorded.

In the age group data area, data relating to an identifier for specifying the age group is recorded.
Data relating to an identifier for identifying the occupation of the customer is recorded in the customer segment data area.

  The transaction pattern data area contains data related to general transaction patterns (statistics on items such as transaction channel, transaction place, transaction time zone, transaction amount, etc.) performed by customers in this region and age group in this transaction type. To be recorded. Here, as the statistical value, a representative value such as an average value or a dispersion range indicating a range in which many transactions are performed is used.

  As shown in FIG. 2D, the attention pattern information storage unit 25 records attention management data 250 regarding patterns (attention patterns) to be noted regarding transactions that may cause problems. The attention management data 250 is recorded when contents to be noted are registered for a transaction having a problem. In the attention management data 250, data regarding attention patterns and coping methods are recorded.

  In the attention pattern data area, data relating to items to be noted in transactions at financial institutions is recorded. In this attention pattern, data related to individual patterns, customer patterns, general patterns, different items in each pattern, payee account attributes to be noted, and keywords to be noted in conversation (caution keywords) are recorded. The Further, as the bank account pattern to be noted, an elapsed period since the opening of the account (the number of days elapsed is less than the standard number of days), a transfer situation (transfer from various individuals), an account with an accident report, and the like are recorded.

  In the coping method data area, data on coping methods (confirmation information) corresponding to the attention pattern is recorded. For example, data on the contents of the interview to be confirmed and advice on how to listen to the customer are recorded. In the present embodiment, data relating to a coping method is recorded in association with a transaction channel.

As a coping method in the case of window transaction, the contents to be heard are output on the display of the window terminal 11.
As a coping method in the case of a transaction using the automatic teller machine 12 or the user terminal 13, the contents to be heard are output on the display of the operator terminal 15. Furthermore, when it is determined that a countermeasure is necessary, the operator terminal 15 calls the handset provided in the automatic teller machine 12, or the user terminal 13 is set to a predetermined telephone number connected to the operator terminal 15. Output a message prompting you to call.

  As shown in FIG. 2 (e), the transaction request management data 260 for the transaction request received from the transaction terminal 10 is recorded in the transaction request storage unit 26. The transaction request management data 260 is recorded when received from the transaction terminal 10. In the transaction request management data 260, data related to transaction date / time, transaction terminal, account code, customer code, transaction type, transaction content, customer status, verification result, and transaction result are recorded.

In the transaction date / time data area, data related to the date and time when the transaction request is received is recorded.
In the transaction terminal data area, data relating to an identifier for identifying the transaction terminal 10 used for the transaction request is recorded. In the present embodiment, data relating to identifiers (terminal numbers and IP addresses) for specifying the window terminal 11, the automatic teller machine 12, and the user terminal 13 of the store of the financial institution are recorded.

In the account code data area, data relating to an identifier for specifying an account used for the transaction is recorded.
In the customer code data area, data relating to an identifier for identifying a customer who conducts the transaction is recorded.

In the transaction type data area, data relating to an identifier for identifying the type of transaction request target is recorded.
In the transaction content data area, data relating to the content of the transaction request (amount, transfer destination, etc.) is recorded.

  In the customer status data area, data relating to the customer status at the time of transaction is recorded. In the present embodiment, biometric information at the time of biometric authentication, gender, age feature quantity, and data on the personal status are recorded as the customer status.

  In the verification result data area, information for specifying the determination result is recorded for the verification results of various verifications. Here, a warning flag is recorded for an item that does not match the individual pattern, the general pattern, the customer pattern, or an item that matches the attention pattern.

  In the transaction result data area, a flag for specifying whether or not there is a problem after the transaction is recorded. When there is a problem report, a problem flag is recorded in this data area as information indicating that there is a problem.

A processing procedure for confirming a customer situation and supporting an appropriate transaction using the system configured as described above will be described with reference to FIGS.
(Overall flow)
First, the overall flow will be described with reference to FIG.
Here, first, the control unit 21 of the customer confirmation support server 20 executes a transaction request process (step S1-1). Specifically, the customer transmits a transaction request to the host system 30 using the transaction terminal 10 (the window terminal 11, the automatic teller machine 12, and the user terminal 13). As will be described later, the transaction request includes data relating to a transaction channel (the window terminal 11, the automatic teller machine 12, and the user terminal 13), customer information, and transaction details (item values corresponding to transaction items). In this case, in the case of dealing with a window, the transaction request includes customer identification information (customer code or account code) for identifying the customer or account input at the window terminal 11 and information on the transaction content. When the automated teller machine 12 is used, the card information (account code) for specifying the cash card inserted into the automated teller machine 12 or the transaction input to the automated teller machine 12 is used in the transaction request. Include information about the content. In the case of using the net by the user terminal 13, the transaction request includes customer identification information (customer code) input to the user terminal 13 and information regarding transaction contents. Here, the host system 30 transmits the transaction request acceptance information to the customer confirmation support server 20 and temporarily holds the transaction. The transaction request acceptance information includes a customer code acquired from the window terminal 11 and the user terminal 13, a customer code specified based on the card information acquired from the automatic teller machine 12, and data relating to transaction details. In this case, the request information acquisition unit 211 of the control unit 21 of the customer confirmation support server 20 acquires the transaction request reception information transmitted to the host system 30. In this case, the confirmation management unit 212 generates transaction request management data 260 in which the transaction date / time, customer code, and transaction details are recorded, and registers the transaction request management data 260 in the transaction request storage unit 26.

  Next, the control part 21 of the customer confirmation support server 20 performs a pattern matching process (step S1-2). Specifically, the confirmation management unit 212 of the control unit 21 performs a confirmation process (store visitor confirmation process, network user confirmation process) according to the transaction channel, as will be described later. In each confirmation process, pattern matching using various patterns (individual patterns, general patterns, customer patterns, attention patterns) is performed. Furthermore, the control part 21 of the customer confirmation support server 20 performs a transfer destination collation process.

  Next, the control part 21 of the customer confirmation support server 20 performs the determination process about whether alerting is required (step S1-3). Specifically, the confirmation management unit 212 of the control unit 21 needs to call attention when a warning flag is recorded in the verification result data area of the transaction request management data 260 as a result of various pattern matching processes. judge.

  When it is determined that attention is required (in the case of “YES” in step S1-3), the control unit 21 of the customer confirmation support server 20 executes a collation result output process (step S1-4). Specifically, the confirmation management unit 212 of the control unit 21 outputs the result of pattern matching. Here, in the case of a window transaction, the result of pattern matching is output on the display of the window terminal 11. In the case of a transaction using the automatic teller machine 12 or the user terminal 13, the result of pattern matching is output to the operator terminal 15. Here, the confirmation management unit 212 outputs the alert information so as to identify items that are common to the pattern and items that are not common in the pattern matching. Here, the person in charge of the financial institution determines whether or not the individual pattern, the general pattern, and the customer pattern need to be handled for items that do not match each pattern. As for the attention pattern, it is determined whether or not the item corresponding to the attention pattern needs to be dealt with.

  Next, the control part 21 of the customer confirmation assistance server 20 performs the determination process about whether a countermeasure is required (step S1-5). Specifically, the person in charge of the financial institution determines whether or not countermeasures are necessary, and inputs countermeasure necessity information at the window terminal 11 or the operator terminal 15. In this case, the confirmation management unit 212 of the control unit 21 acquires information (necessary action information or unnecessary action information) for identifying whether the action is necessary or unnecessary from the window terminal 11 or the operator terminal 15.

  Here, if necessary information is acquired from the window terminal 11 or the operator terminal 15 and it is determined that the countermeasure is necessary (“YES” in step S1-5), the control unit 21 of the customer confirmation support server 20 performs the hearing. Processing is executed (step S1-6). Specifically, the confirmation management unit 212 of the control unit 21 acquires a coping method corresponding to the collation result from the attention pattern information storage unit 25 and outputs information necessary for the hearing to the window terminal 11 and the operator terminal 15. To do. For example, when there is an inconsistent item in an individual pattern, a general pattern, or a customer pattern, information related to a question for confirming the person is output. When there is a matching item in the attention pattern, information related to a question for confirming the transaction is output. In this case, the person in charge of the financial institution conducts hearing using information output to the window terminal 11 and the operator terminal 15. For example, in the case of a window transaction, an interview is performed based on the contents output on the display of the window terminal 11. Further, in the case of a transaction using the automatic teller machine 12, the operator terminal 15 calls the handset provided in the automatic teller machine 12, and based on the contents output on the display of the operator terminal 15. Hear. Further, in the case of a transaction using the user terminal 13, based on the content output on the display of the operator terminal 15 when a call is received from the customer according to the message output to the user terminal 13. Hear. Then, the person in charge of the financial institution registers information (for example, family status or event status) obtained through the hearing in the customer management data 230 of the customer in the customer pattern information storage unit 23 as necessary.

  Next, the control part 21 of the customer confirmation support server 20 performs the determination process about whether there is any problem (step S1-7). Specifically, the person in charge of the financial institution determines whether there is a problem based on the result of the hearing, and inputs this determination result to the window terminal 11 and the operator terminal 15. And the confirmation management part 212 of the control part 21 determines whether there is any problem based on the judgment result input in the window terminal 11 or the operator terminal 15.

  When the determination result that there is a problem is acquired (in the case of “NO” in step S1-7), the control unit 21 of the customer confirmation support server 20 executes a transaction hold process (step S1-8). Specifically, the confirmation management unit 212 of the control unit 21 holds the transaction as it is. In this case, the host system 30 continues to hold the transaction request. And the person in charge of a financial institution confirms further about a transaction with respect to a customer.

  On the other hand, when the determination result that there is no problem is acquired (in the case of “YES” in step S1-7), the control unit 21 of the customer confirmation support server 20 executes a transaction process (step S1-9). Specifically, the confirmation management unit 212 of the control unit 21 notifies the host system 30 that there is no problem as a result of the identity verification. In this case, the host system 30 executes processing for the pending transaction request.

  Note that when it is determined that alerting is not necessary (in the case of “NO” in step S1-3), or when it is determined that no action is required after acquiring unnecessary information from the window terminal 11 or the operator terminal 15 (step S1) In the case of “NO” in −5), the control unit 21 of the customer confirmation support server 20 executes the transaction process (step S1-9). In this case, the host system 30 executes the pending transaction request.

  After the transaction is completed, the customer confirmation support server 20 performs the following processing. Here, if it is later determined that there is a problem with the transaction, the customer reports the problem to the financial institution. In this case, in the transaction request storage unit 26, the transaction request management data 260 for the transaction having a problem is specified, and a problem flag is recorded in the transaction result data area of the transaction request management data 260. And the control part 21 of the customer confirmation assistance server 20 performs the determination process about whether there was any problem after dealings (step S1-10). Specifically, when the problem flag is recorded in the transaction result data area of the transaction request management data 260, the confirmation management unit 212 of the control unit 21 determines that there is a problem with the transaction. On the other hand, if a problem flag is not recorded in the transaction result data area of the transaction request management data 260 even after a predetermined period has elapsed from the transaction date and time, the confirmation management unit 212 of the control unit 21 indicates that there is no problem with the transaction. judge.

  When it is determined that there is no problem after the transaction (in the case of “YES” in step S1-10), the control unit 21 of the customer confirmation support server 20 executes the update process of the individual pattern, the customer pattern, and the general pattern (step S1). -11). Specifically, the confirmation management unit 212 of the control unit 21 updates the individual pattern recorded in the individual pattern information storage unit 22 and the customer pattern recorded in the customer pattern information storage unit 23 based on the transaction content. . Further, the statistical value is recalculated based on the area where the transaction is performed and the customer attribute, and the general pattern recorded in the general pattern information storage unit 24 is updated.

  On the other hand, when it is determined that there is a problem after the transaction (in the case of “NO” in step S1-10), the control unit 21 of the customer confirmation support server 20 executes a caution pattern update process (step S1-12). Specifically, the confirmation management unit 212 of the control unit 21 updates the caution pattern recorded in the caution pattern information storage unit 25 based on the transaction content for which there was a problem report. Here, the confirmation management unit 212 searches the attention pattern information storage unit 25 for a attention pattern having a common item value for the problematic transaction. When a caution pattern having a common item value is already recorded in the caution pattern information storage unit 25, the confirmation management unit 212 updates the item value of the caution pattern based on the transaction content in which the problem report has been made. On the other hand, when a caution pattern having a common item value is not recorded in the caution pattern information storage unit 25, the confirmation management unit 212 uses a transaction content for which a problem report has been issued to make a note about a new caution pattern. Management data 250 is generated and recorded in the attention pattern information storage unit 25.

(Visitor confirmation process)
Next, the store visitor confirmation process corresponding to the window terminal 11 and the automatic teller machine 12 will be described with reference to FIG. This process is executed when a transaction is performed using the window terminal 11 or the automatic teller machine 12.

  First, the control unit 21 of the customer confirmation support server 20 executes store visitor information acquisition processing (step S2-1). Specifically, the confirmation management unit 212 of the control unit 21 specifies a store visit location and a store visit time. In this embodiment, the store where the window terminal 11 and the automatic teller machine 12 to which the transaction request is transmitted is installed is specified as the store visit location. The date and time when the transaction request is received is specified as the store visit time.

  Next, the control unit 21 of the customer confirmation support server 20 executes a store visitor image acquisition process (step S2-2). Specifically, the confirmation management unit 212 of the control unit 21 acquires a store visitor image obtained by photographing a store visitor at the store where the contact terminal 11 is installed. For example, an in-store camera is used at the window. When the automatic teller machine 12 is used, the camera of the automatic teller machine 12 is used. And the confirmation management part 212 calculates the feature-value about a face image, sex, and age based on a visitor image.

  Next, the control part 21 of the customer confirmation support server 20 performs the account holder specific process (step S2-3). Specifically, when the automatic teller machine 12 is used, the confirmation management unit 212 of the control unit 21 specifies the account used for the transaction by the inserted cash card. In the case of using the window, the account used for the transaction is specified based on the account information input to the window terminal 11 by the person in charge of the financial institution. Then, the confirmation management unit 212 uses the account information storage unit 33 of the host system 30 to specify the customer code of the account holder of the account used for the transaction.

  Next, the control unit 21 of the customer confirmation support server 20 executes transaction content acquisition processing (step S2-4). Specifically, the confirmation management unit 212 of the control unit 21 acquires transaction information input to the automatic teller machine 12 or the window terminal 11.

  Next, the control part 21 of the customer confirmation support server 20 performs the determination process about whether there is a change of a life pattern (step S2-5). Specifically, the confirmation management unit 212 of the control unit 21 acquires the customer management data 230 in which the customer code of the specified account holder is recorded from the customer pattern information storage unit 23. Then, the confirmation management unit 212 confirms whether there is a life event recorded in the event status data area of the customer management data 230. Here, when a specific type of life event that has occurred within a predetermined period from the current date and time is recorded, it is determined that there is a change in the life pattern.

  When it is determined that there is a change in the life pattern (in the case of “YES” in step S2-5), the control unit 21 of the customer confirmation support server 20 executes a general pattern matching process (step S2-6). Specifically, the confirmation management unit 212 of the control unit 21 performs a general pattern matching process described later.

  On the other hand, when it is determined that there is no change in the life pattern (in the case of “NO” in step S2-5), the control unit 21 of the customer confirmation support server 20 executes a process of searching for the account holder's usage history (step) S2-7). Specifically, the confirmation management unit 212 of the control unit 21 searches the individual pattern information storage unit 22 for the individual pattern data 220 in which the account code of this account is recorded.

  Next, the control part 21 of the customer confirmation support server 20 performs the determination process about whether there exists accumulation | storage of a use history (step S2-8). Specifically, the confirmation management unit 212 of the control unit 21 stores the usage history when the individual pattern information storage unit 22 records the individual pattern data 220 in which the account code of this account is recorded. Judge that there is.

  If the individual pattern data 220 in which the account code of this account is recorded cannot be extracted and it is determined that the usage history is not accumulated (in the case of “NO” in step S2-8), the control unit 21 of the customer confirmation support server 20 Performs a general pattern matching process to be described later (step S2-6).

  On the other hand, when it is determined that the usage history is accumulated (in the case of “YES” in step S2-8), the control unit 21 of the customer confirmation support server 20 executes a process of specifying past feature information (step S2- 9). Specifically, the confirmation management unit 212 of the control unit 21 extracts the customer management data 230 in which the customer code of the account holder is recorded from the customer pattern information storage unit 23. Then, the confirmation management unit 212 obtains the feature amount for the customer pattern (biometric authentication, gender, age, personal status (face expression)) recorded in the customer management data 230 in which the transaction type of the current transaction request is recorded. Identify.

  Next, the control unit 21 of the customer confirmation support server 20 executes a determination process as to whether or not the features match (step S2-10). Specifically, the confirmation management unit 212 of the control unit 21 compares the feature amount specified in the customer management data 230 with the feature amount calculated from the store visitor image. Here, the confirmation management unit 212 determines that the features match if the difference between the feature amounts is equal to or less than a predetermined allowable value.

When it is determined that the features match (in the case of “YES” in step S2-10), the control unit 21 of the customer confirmation support server 20 executes the identity verification process (step S2-11). This process will be described later with reference to FIG.
Furthermore, the control part 21 of the customer confirmation support server 20 performs an individual pattern collation process (step S2-12). This process will be described later with reference to FIG.

  On the other hand, when it is determined that the features do not match (in the case of “NO” in step S2-10), the control unit 21 of the customer confirmation support server 20 executes a process for acquiring the family information of the account holder (step S2- 13). Specifically, the confirmation management unit 212 of the control unit 21 acquires information about the family status of the customer from the customer management data 230 of the customer pattern information storage unit 23.

  Next, the control part 21 of the customer confirmation support server 20 performs a transaction attribute specific process (step S2-14). Specifically, the confirmation management unit 212 of the control unit 21 uses a known pattern recognition technique to predict gender and age based on the store visitor image.

  Next, the control part 21 of the customer confirmation assistance server 20 performs the determination process about whether there is a possibility of a proxy (step S2-15). Specifically, the confirmation management unit 212 of the control unit 21 displays the agent when the family status data area of the customer management data 230 includes a family that matches the gender and age predicted based on the store visitor image. It is determined that there is a possibility.

  When it is determined that there is a possibility of an agent (in the case of “YES” in step S2-15), the control unit 21 of the customer confirmation support server 20 executes an individual pattern matching process described later (step S2-12). .

  On the other hand, when it is determined that there is no possibility of the agent (in the case of “NO” in step S2-15), the control unit 21 of the customer confirmation support server 20 executes a warning flag recording process (step S2-16). ). Specifically, the confirmation management unit 212 of the control unit 21 records an alert flag in association with the verification result of the visitor confirmation in the transaction request management data 260 of the transaction request storage unit 26.

(Internet user confirmation processing)
Next, the network user confirmation process corresponding to the user terminal 13 will be described with reference to FIG. Here, the case where the web camera is provided in the user terminal 13 and the case where the web camera is not provided are assumed. If a web camera is provided, the user's image taken with this web camera is transmitted to the customer confirmation support server 20.

  Here, first, the control unit 21 of the customer confirmation support server 20 executes accessor information acquisition processing (step S3-1). Specifically, when performing online transactions, the user accesses the host system 30 using the user terminal 13. In this case, the host system 30 acquires the IP address or the like of the user terminal 13 as access source information. Then, the confirmation management unit 212 of the control unit 21 of the customer confirmation support server 20 acquires access source information from the host system 30. Furthermore, the confirmation management unit 212 specifies the current time as the transaction time, generates transaction request management data 260 in which the transaction time is recorded, and records it in the transaction request storage unit 26.

  Next, the control part 21 of the customer confirmation support server 20 performs the account holder specific process (step S3-2). Specifically, the host system 30 outputs a login screen to the user terminal 13 and acquires login information (customer code and password) input on the login screen. Based on the login information, the host system 30 performs login authentication as to whether or not there is a registration in the customer information storage unit 32. The host system 30 specifies an accessor (account holder) based on the login authentication result. In this case, the confirmation management unit 212 of the control unit 21 of the customer confirmation support server 20 acquires a customer code from the host system 30.

  Next, the control unit 21 of the customer confirmation support server 20 executes transaction content acquisition processing (step S3-3). Specifically, the host system 30 outputs a transaction screen to the user terminal 13 and acquires the transaction content input on the transaction screen. In this case, the confirmation management unit 212 of the control unit 21 acquires information about transaction details from the host system 30.

Next, the control part 21 of the customer confirmation assistance server 20 performs the determination process about whether there is a change of a life pattern similarly to step S2-5 (step S3-4).
When it is determined that there is a change in the life pattern (in the case of “YES” in step S3-4), the control unit 21 of the customer confirmation support server 20 executes the general pattern matching process as in step S2-6 ( Step S3-5).

  On the other hand, when it is determined that there is no change in the life pattern (“NO” in step S3-4), the control unit 21 of the customer confirmation support server 20 uses the account holder's usage history in the same manner as in step S2-7. The search process is executed (step S3-6).

Next, similarly to step S2-8, the control unit 21 of the customer confirmation support server 20 executes a determination process as to whether or not there is usage history accumulation (step S3-7).
When it is determined that the usage history is not accumulated (in the case of “NO” in step S3-7), the control unit 21 of the customer confirmation support server 20 executes the general pattern matching process as in step S2-6 ( Step S3-5).

  On the other hand, when it is determined that the usage history is accumulated (in the case of “YES” in step S3-7), the control unit 21 of the customer confirmation support server 20 executes a determination process as to whether or not the web camera is used. (Step S3-8). Specifically, the confirmation management unit 212 of the control unit 21 confirms whether or not the web camera can be used with respect to the user terminal 13, and requests the start of the camera when the web camera can be used. Send the shooting confirmation screen. Here, when the web camera can be used, the user activates the web camera of the user terminal 13. In this case, the user terminal 13 transmits the user image taken by the web camera to the customer confirmation support server 20. When the web camera cannot be used, the user inputs “unusable” on the shooting confirmation screen. In this case, the user terminal 13 transmits an unusable message to the customer confirmation support server 20. Then, the confirmation management unit 212 determines that the web camera can be used when the user image is acquired, and determines that it cannot be used when the unavailable message is acquired.

  Here, when it is determined that the web camera is not used (in the case of “NO” in step S3-8), the individual pattern matching process described later is executed as in step S2-12 (step S3-13). .

  On the other hand, when it is determined that the web camera is used (“YES” in step S3-8), the control unit 21 of the customer confirmation support server 20 executes an image acquisition process (step S3-9). Specifically, the confirmation management unit 212 of the control unit 21 acquires an image taken by the web camera of the user terminal 13. And the confirmation management part 212 calculates the feature-value about a facial expression, sex, and age based on the image image | photographed with the web camera of the user terminal 13. FIG.

  Next, the control unit 21 of the customer confirmation support server 20 executes past feature information specifying processing in the same manner as in step S2-9 (step S3-10). Next, the control part 21 of the customer confirmation support server 20 performs the determination process about whether the characteristic is the same as step S2-10 (step S3-11).

  When it is determined that the features match (in the case of “YES” in step S3-11), the control unit 21 of the customer confirmation support server 20 executes the personal status matching process in the same manner as in step S2-11 (step S3- 12).

Next, the control part 21 of the customer confirmation support server 20 performs an individual pattern collation process similarly to step S2-12 (step S3-13).
On the other hand, if it is determined that the features do not match (in the case of “NO” in step S3-11), the control unit 21 of the customer confirmation support server 20 acquires the family information of the account holder as in step S2-13. Processing is executed (step S3-14).

  Next, the control part 21 of the customer confirmation assistance server 20 performs the transaction attribute specific process similarly to step S2-14 (step S3-15). Here, the user attributes (gender, age) are specified using an image acquired from the user terminal 13 instead of the store visitor image.

Next, the control part 21 of the customer confirmation support server 20 performs the determination process about whether there is a possibility of a proxy similarly to step S2-15 (step S3-16).
When it is determined that there is a possibility of an agent (in the case of “YES” in step S3-16), the control unit 21 of the customer confirmation support server 20 executes the individual pattern matching process as in step S2-12. (Step S3-13).

  On the other hand, when it is determined that there is no possibility of the agent (in the case of “NO” in step S3-16), the control unit 21 of the customer confirmation support server 20 performs the alert flag recording process in the same manner as in step S2-16. Is executed (step S3-17).

(General pattern matching process)
Next, the general pattern matching process will be described with reference to FIG.
Here, first, the control unit 21 of the customer confirmation support server 20 executes an account holder attribute acquisition process (step S4-1). Specifically, the general verification unit 213 of the control unit 21 acquires information on customer attributes (address, age, occupation, etc.) from the customer information storage unit 32 of the host system 30.

  Next, the control unit 21 of the customer confirmation support server 20 executes a general pattern specifying process corresponding to the attribute (step S4-2). Specifically, the general matching unit 213 of the control unit 21 acquires a general pattern corresponding to the customer attribute from the general pattern information storage unit 24. Specifically, a general pattern corresponding to the area to which the address belongs, the age group to which the age belongs, and the customer group to which the occupation belongs is acquired.

  Next, the control part 21 of the customer confirmation assistance server 20 performs a comparison process with the general pattern corresponding to an attribute (step S4-3). Specifically, the general verification unit 213 of the control unit 21 compares the item value of the transaction type included in the transaction content with the statistical value included in the general pattern.

  Next, the control part 21 of the customer confirmation assistance server 20 performs the determination process about whether there exists similarity (step S4-4). Specifically, the general matching unit 213 of the control unit 21 has similarity when the item value included in the transaction content is included in a statistical value (a predetermined range or dispersion range from the representative value) included in the general pattern. Judge that there is.

  When it determines with there being no similarity (in the case of "NO" in step S4-4), the control part 21 of the customer confirmation assistance server 20 performs an alert flag recording process (step S4-5). Specifically, the general verification unit 213 of the control unit 21 records an alert flag in association with the verification result of the visitor confirmation in the transaction request management data 260 of the transaction request storage unit 26.

  On the other hand, if it is determined that there is similarity (in the case of “YES” in step S4-4), the control unit 21 of the customer confirmation support server 20 skips the alert flag recording process (step S4-5).

  Next, the control part 21 of the customer confirmation support server 20 performs a comparison process with an attention pattern (step S4-6). Specifically, the general verification unit 213 of the control unit 21 acquires a caution pattern corresponding to the general pattern recorded in the caution management data 250 of the caution pattern information storage unit 25. Then, the general matching unit 213 compares the item value corresponding to the transaction content with the attention pattern.

  Next, the control part 21 of the customer confirmation assistance server 20 performs the determination process about whether there exists similarity (step S4-7). Specifically, the general verification unit 213 of the control unit 21 determines that there is similarity when there is an item value included in the transaction content that corresponds to the attention pattern.

  When it is determined that there is similarity (in the case of “YES” in step S4-7), the control unit 21 of the customer confirmation support server 20 executes the alert flag recording process as in step S4-5 (step S4-5). S4-8). On the other hand, when it determines with there being no similarity (in the case of "NO" in step S4-7), the control part 21 of the customer confirmation assistance server 20 skips an alert flag recording process (step S4-8).

(Person status check process)
Next, the personal status collation process will be described with reference to FIG.
First, the control part 21 of the customer confirmation support server 20 performs the determination process about whether a thermo image is acquirable (step S5-1). Specifically, the state collation unit 215 of the control unit 21 transmits a thermo image acquisition permission confirmation to the transaction terminal 10. In this case, when the transaction terminal 10 includes a thermography device, the transaction terminal 10 returns possible information regarding acquisition of the thermo image. On the other hand, when the thermographic device is not provided, the transaction terminal 10 sends back information indicating that the thermo image is acquired.

  When it is determined that the thermo image can be acquired by acquiring the possibility information (in the case of “YES” in step S5-1), the control unit 21 of the customer confirmation support server 20 executes a thermo image acquisition process ( Step S5-2). Specifically, the state matching unit 215 of the control unit 21 acquires a thermo image of the transaction from the thermography device of the transaction terminal 10. And the state collation part 215 calculates a body temperature feature-value based on a thermo image.

  Next, the control part 21 of the customer confirmation support server 20 performs the identification process of the past thermo image information of the account holder (step S5-3). Specifically, the state verification unit 215 of the control unit 21 acquires the customer management data 230 in which the customer code is recorded from the customer pattern information storage unit 23. Then, the state matching unit 215 acquires body temperature information recorded as the personal situation in the customer pattern data area of the customer management data 230 in association with the current transaction type.

  Next, the control part 21 of the customer confirmation assistance server 20 performs the determination process about whether a body temperature distribution is an allowable range (step S5-4). Specifically, the state verification unit 215 of the control unit 21 compares the body temperature information (body temperature feature amount) specified in the customer management data 230 with the body temperature feature amount calculated from the thermo image. Then, the confirmation management unit 212 determines that the body temperature distribution is within the allowable range when the difference between the two feature amounts is equal to or less than a predetermined allowable value.

  When it is determined that the body temperature distribution is not within the allowable range (in the case of “NO” in step S5-4), the control unit 21 of the customer confirmation support server 20 executes a warning flag recording process (step S5-5). Specifically, the state verification unit 215 of the control unit 21 records an alert flag in association with the verification result of the visitor confirmation in the transaction request management data 260 of the transaction request storage unit 26.

  On the other hand, when it is determined that the body temperature distribution is within the allowable range (in the case of “YES” in step S5-4), the control unit 21 of the customer confirmation support server 20 skips the alert flag recording process (step S5-5).

  If it is determined that the thermo image cannot be acquired (“NO” in step S5-1), the control unit 21 of the customer confirmation support server 20 skips the processes in steps S5-2 to S5-5.

  Next, the control unit 21 of the customer confirmation support server 20 executes behavior information acquisition processing (step S5-6). Specifically, the state matching unit 215 of the control unit 21 calculates a behavior feature amount based on an image acquired from the transaction terminal 10. Here, as the feature amount of the behavior, the amount of change in motion is calculated for the facial expression of the trader and how to move the body. Moreover, in the case of the transaction terminal 10 provided with the microphone, the state collation part 215 acquires the audio | voice collected with the microphone. And the state collation part 215 specifies the feature-value of a trader's voice.

  Next, the control part 21 of the customer confirmation support server 20 performs the identification process of the past behavior information of the account holder (step S5-7). Specifically, the state verification unit 215 of the control unit 21 relates to the behavior recorded as the personal situation in the customer pattern data area of the customer management data 230 of the customer pattern information storage unit 23 in association with the current transaction type. Get information about feature quantities.

  Next, the control part 21 of the customer confirmation assistance server 20 performs the determination process about whether there exists similarity (step S5-8). Specifically, the state matching unit 215 of the control unit 21 records the image and audio feature quantities acquired from the transaction terminal 10 in association with the current transaction type in the customer pattern data area of the customer management data 230. It is determined that there is similarity when it is included in a predetermined range from the feature amount of the behavior.

  When it is determined that there is no similarity (in the case of “NO” in step S5-8), the control unit 21 of the customer confirmation support server 20 executes a warning flag recording process as in step S5-5 (step S5-5). S5-9).

  On the other hand, when it is determined that there is similarity (in the case of “YES” in step S5-8), the control unit 21 of the customer confirmation support server 20 skips the alert flag recording process (step S5-9).

  Next, the control part 21 of the customer confirmation support server 20 performs the confirmation process of the attention pattern keyword (step S5-10). Specifically, the state verification unit 215 of the control unit 21 performs voice recognition of the voice acquired from the transaction terminal 10 provided with a microphone. Then, the state collation unit 215 confirms whether or not the attention keyword recorded in the attention management data 250 of the attention pattern information storage unit 25 is included in the voice.

  Next, the control part 21 of the customer confirmation assistance server 20 performs the determination process about whether there is any problem (step S5-11). Specifically, the state matching unit 215 of the control unit 21 determines that there is a problem in behavior when the caution keyword recorded in the caution management data 250 is included in the voice.

  When it is determined that there is a problem (in the case of “NO” in step S5-11), the control unit 21 of the customer confirmation support server 20 executes a warning flag recording process as in step S5-5 (step S5). -12).

  On the other hand, when it is determined that there is no problem (in the case of “YES” in step S5-11), the control unit 21 of the customer confirmation support server 20 skips the alert flag recording process (step S5-12).

(Individual pattern matching process)
Next, the individual pattern matching process will be described with reference to FIG.
First, the control unit 21 of the customer confirmation support server 20 executes an attribute holder attribute acquisition process (step S6-1). Specifically, the individual verification unit 214 of the control unit 21 acquires information on customer attributes (address, age, occupation, etc.) from the customer information storage unit 32 of the host system 30.

  Next, the control unit 21 of the customer confirmation support server 20 executes a transaction pattern identification process for the account holder (step S6-2). Specifically, the individual verification unit 214 of the control unit 21 acquires the individual pattern data 220 in which the account code used for the transaction is recorded from the individual pattern information storage unit 22. Then, the individual verification unit 214 acquires the individual pattern recorded in the transaction pattern data area of the individual pattern data 220.

  Next, the control part 21 of the customer confirmation support server 20 performs a comparison process with the transaction pattern of an account holder (step S6-3). Specifically, the individual verification unit 214 of the control unit 21 compares the transaction channel, the transaction place, the transaction time zone, the transaction amount, the business partner in the current transaction, and the individual pattern of the individual pattern data 220.

  Next, the control part 21 of the customer confirmation assistance server 20 performs the determination process about whether there exists similarity (step S6-4). Specifically, the individual verification unit 214 of the control unit 21 includes a transaction channel, a transaction place, a transaction time zone, a transaction amount, and a customer in the current transaction within a pattern included in the individual pattern (dispersion range from representative value). If it is included, it is determined that there is similarity.

  When it is determined that there is similarity (in the case of “YES” in step S6-4), the control unit 21 of the customer confirmation support server 20 executes a determination process as to whether there has been a problem in the past transaction ( Step S6-5). Specifically, the individual verification unit 214 of the control unit 21 searches for past transaction request management data 260 having a common account code or customer code. Furthermore, the individual verification unit 214 searches the transaction request management data 260 having the same transaction type, transaction content, and customer status for data recorded as having a problem in the transaction result data area. When the transaction request management data 260 in which the problem flag is recorded is extracted, it is determined that there was a problem in the past transaction.

  When it is determined that there has been a problem in the past transaction (in the case of “YES” in step S6-5), the control unit 21 of the customer confirmation support server 20 executes a warning flag recording process (step S6-6). Specifically, the individual verification unit 214 of the control unit 21 records an alert flag in association with the verification result of the transaction confirmation in the transaction request management data 260 of the transaction request storage unit 26.

  On the other hand, when it is determined that there was no problem in the past transaction (in the case of “NO” in step S6-5), the control unit 21 of the customer confirmation support server 20 skips the alert flag recording process (step S6-6). To do.

  Moreover, also when it determines with there being no similarity (in the case of "NO" in step S6-4), the control part 21 of the customer confirmation assistance server 20 performs a warning flag recording process (step S6-6). .

  Next, the control part 21 of the customer confirmation support server 20 performs a comparison process with an attention pattern (step S6-7). Specifically, the individual verification unit 214 of the control unit 21 acquires a caution pattern corresponding to the individual pattern recorded in the caution management data 250 of the caution pattern information storage unit 25. Then, the individual verification unit 214 compares the item value corresponding to the transaction content with the attention pattern.

Next, the control part 21 of the customer confirmation assistance server 20 performs the determination process about whether there exists similarity similarly to step S4-7 (step S6-8).
When it is determined that there is similarity (in the case of “YES” in step S6-8), the control unit 21 of the customer confirmation support server 20 executes a warning flag recording process in the same manner as in step S6-6 (step S6-6). S6-9). On the other hand, when it determines with there being no similarity (in the case of "NO" in step S6-8), the control part 21 of the customer confirmation assistance server 20 skips an alert flag recording process (step S6-9).

(Bank transfer verification process)
Next, the transfer destination verification process will be described with reference to FIG. Here, in the transfer destination account, the transfer destination account is evaluated using information such as the transfer status, transfer frequency, and account opening period from others.

  First, the control part 21 of the customer confirmation support server 20 performs the determination process about whether there exists a transfer performance in the same transfer destination (step S7-1). Specifically, the transfer destination verification unit 216 of the control unit 21 extracts past transaction request management data 260 in which the account code and customer code in the current transaction are recorded from the transaction request storage unit 26. In the extracted transaction request management data 260, when there is data in which the account code of the transfer destination in the current transaction is recorded, it is determined that the same transfer destination has a transfer record.

  When it is determined that there is a transfer record at the same transfer destination (in the case of “YES” in step S7-1), the control unit 21 of the customer confirmation support server 20 executes a determination process as to whether there is no problem (step). S7-2). Specifically, the transfer destination verification unit 216 of the control unit 21 confirms the presence / absence of the transaction request management data 260 in which a problem flag is recorded in the transaction result data area. If there is transaction request management data 260 in which a problem flag is recorded, it is determined that there is a problem.

  When it is determined that there is a problem (in the case of “NO” in step S7-2), the control unit 21 of the customer confirmation support server 20 executes a warning flag recording process (step S7-3). Specifically, the transfer destination verification unit 216 of the control unit 21 records an alert flag in association with the verification result of the transfer destination confirmation in the transaction request management data 260 of the transaction request storage unit 26. On the other hand, if it is determined that there is no problem (“YES” in step S7-2), the control unit 21 of the customer confirmation support server 20 skips the alert flag recording process (step S7-3). Further, when it is determined that there is no transfer record at the same transfer destination (in the case of “NO” in step S7-1), the processes of steps S7-2 and S7-3 are skipped.

  Next, the control part 21 of the customer confirmation support server 20 performs the identification process of the account condition of a transfer destination (step S7-4). Specifically, the transfer destination verification unit 216 of the control unit 21 acquires the account information of the transfer destination account from the account information storage unit 33 of the host system 30. Here, the transfer destination collation unit 216 acquires information on the opening date of the transfer destination account and the accident report. Furthermore, the transfer destination verification unit 216 calculates the usage frequency based on the transaction history of this account.

  Next, the control part 21 of the customer confirmation support server 20 performs a comparison process with an attention pattern (step S7-5). Specifically, the transfer destination verification unit 216 of the control unit 21 acquires a caution pattern related to the transfer destination account recorded in the caution management data 250 of the caution pattern information storage unit 25. Then, the transfer destination verification unit 216 compares the current transfer destination account opening time, usage frequency, and attention pattern. Further, the transfer destination verification unit 216 checks whether there is an accident report.

  Next, the control part 21 of the customer confirmation assistance server 20 performs the determination process about whether there exists similarity (step S7-6). Specifically, the transfer destination verification unit 216 of the control unit 21 determines that there is similarity when the current transfer destination account corresponds to the attention pattern.

  When it is determined that there is a similarity (in the case of “YES” in step S7-6), the control unit 21 of the customer confirmation support server 20 executes a warning flag recording process in the same manner as in step S7-3 (step S7-3). S7-7). On the other hand, when it determines with there being no problem (in the case of "NO" in step S7-6), the control part 21 of the customer confirmation assistance server 20 skips an alert flag recording process (step S7-7).

According to this embodiment, the following effects can be obtained.
(1) In this embodiment, the control part 21 of the customer confirmation assistance server 20 performs a pattern matching process (step S1-2). And when it determines with alerting required (in the case of "YES" in step S1-3), the control part 21 of the customer confirmation assistance server 20 performs the output process of a collation result (step S1-4). Here, attention is given so that items common to pattern matching and items not common can be identified. Thereby, it is possible to examine the response while checking the items to be noted.

  Also, if necessary information is acquired from the window terminal 11 or the operator terminal 15 and it is determined that the countermeasure is necessary (“YES” in step S1-5), the control unit 21 of the customer confirmation support server 20 performs the hearing process. Is executed (step S1-6). For example, when there is an inconsistent item in an individual pattern, a general pattern, or a customer pattern, information related to a question for confirming the person is output. When there is a matching item in the attention pattern, information related to a question for confirming the transaction is output. Thereby, the person in charge of the financial institution can perform hearing using the information output to the window terminal 11 and the operator terminal 15.

  And when it determines with there being a problem (in the case of "NO" in step S1-7), the control part 21 of the customer confirmation assistance server 20 performs a transaction suspension process (step S1-8). Thereby, transactions that may cause problems can be suspended.

  (2) In this embodiment, when it is determined that there is no problem after the transaction (in the case of “YES” in step S1-10), the control unit 21 of the customer confirmation support server 20 determines the individual pattern, the customer pattern, and the general pattern. The update process is executed (step S1-11). On the other hand, when it is determined that there is a problem after the transaction (in the case of “NO” in step S1-10), the control unit 21 of the customer confirmation support server 20 executes a caution pattern update process (step S1-12). Thereby, a pattern can be updated corresponding to a situation change.

  (3) In this embodiment, the control part 21 of the customer confirmation assistance server 20 performs the determination process about whether there is a change of a life pattern (step S2-5, S3-4). When it is determined that there is a life pattern change (in the case of “YES” in steps S2-5 and S3-4), the control unit 21 of the customer confirmation support server 20 executes a general pattern matching process (step S2-6). , S3-5). Thereby, when there is a change in the situation such as marriage or moving, pattern matching can be performed using the general pattern without sticking to the individual patterns so far.

  (4) In this embodiment, the control part 21 of the customer confirmation assistance server 20 performs the determination process about whether there exists accumulation | storage of a use history (step S2-8, S3-7). When it is determined that there is no usage history accumulation (NO in steps S2-8 and S3-7), the control unit 21 of the customer confirmation support server 20 executes a general pattern matching process (step S2-6). , S3-5). Thereby, even when there is no accumulation of this customer pattern, it is possible to perform collation using the general pattern.

  (5) In the present embodiment, the control unit 21 of the customer confirmation support server 20 executes past feature information specifying processing (steps S2-9 and S3-10). Next, the control unit 21 of the customer confirmation support server 20 executes a determination process as to whether the features match (steps S2-10 and S3-11). When it is determined that the features match (in the case of “YES” in steps S2-10 and S3-11), the control unit 21 of the customer confirmation support server 20 executes the personal status check process (steps S2-11 and S3-11). 12). Thereby, it is possible to perform verification for detecting a transaction that may cause a problem in consideration of the person's state.

  (6) In this embodiment, when it is determined that the features do not match (in the case of “NO” in steps S2-10 and S3-11), the control unit 21 of the customer confirmation support server 20 determines the family of the account holder Information acquisition processing is executed (steps S2-13 and S3-14). Next, the control part 21 of the customer confirmation support server 20 performs a transaction attribute specific process (steps S2-14 and S3-15). Thereby, even if it is not the account holder himself / herself, verification can be performed in consideration of the family structure and the like.

  (7) In the general pattern matching process of the present embodiment, the control unit 21 of the customer confirmation support server 20 executes an account holder attribute acquisition process (step S4-1). Next, the control part 21 of the customer confirmation assistance server 20 performs a comparison process with the general pattern corresponding to an attribute (step S4-3). And the control part 21 of the customer confirmation assistance server 20 performs the determination process about whether there exists similarity (step S4-4). Thereby, in consideration of the transaction mode according to the customer attribute, it is possible to perform verification for detecting a transaction that may cause a problem.

  (8) In the identity verification process of the present embodiment, when it is determined that a thermo image can be acquired (in the case of “YES” in step S5-1), the control unit 21 of the customer confirmation support server 20 An acquisition process is executed (step S5-2). Next, the control part 21 of the customer confirmation assistance server 20 performs the determination process about whether a body temperature distribution is an allowable range (step S5-4). When it is determined that the body temperature distribution is not within the allowable range (in the case of “NO” in step S5-4), the control unit 21 of the customer confirmation support server 20 executes a warning flag recording process (step S5-5). Thereby, the verification for detecting a transaction that may cause a problem in consideration of the physical condition of the transaction can be performed.

  (9) In the personal status verification process of the present embodiment, the control unit 21 of the customer confirmation support server 20 executes a behavior information acquisition process (step S5-6). Next, the control part 21 of the customer confirmation support server 20 performs the identification process of the past behavior information of the account holder (step S5-7). When it is determined that there is no similarity in the feature amount of the behavior (in the case of “NO” in step S5-8), the control unit 21 of the customer confirmation support server 20 executes a warning flag recording process (step S5-9). ). Thereby, it is possible to detect a transaction that may cause a problem in consideration of the behavior of the trader.

  (10) In the personal status collation process of the present embodiment, the control unit 21 of the customer confirmation support server 20 executes a confirmation process of a keyword for a caution pattern (step S5-10). When it is determined that there is a problem (in the case of “NO” in step S5-11), the control unit 21 of the customer confirmation support server 20 executes a warning flag recording process (step S5-12). Thereby, it is possible to detect a transaction that may cause a problem based on the keyword included in the voice of the customer.

(11) In the individual pattern matching process of the present embodiment, the control unit 21 of the customer confirmation support server 20 executes a comparison process with the transaction pattern of the account holder (step S6-3), and whether there is similarity. The determination process about whether is performed (step S6-4). When it is determined that there is no similarity (in the case of “NO” in step S6-4), the control unit 21 of the customer confirmation support server 20 executes a warning flag recording process (step S6-6). Thereby, when it differs from a customer's normal transaction form, it can alert to the transaction which may cause a problem.
If it is determined that there has been a problem in the past transaction (“YES” in step S6-5), the control unit 21 of the customer confirmation support server 20 executes a warning flag recording process (step S6-6). ). Thereby, it is possible to call attention to transactions that may cause problems in consideration of transaction results.

  (12) In the transfer destination verification process of the present embodiment, when it is determined that there is a transfer record at the same transfer destination (“YES” in step S7-1), the control unit 21 of the customer confirmation support server 20 A determination process as to whether or not there is a problem is executed (step S7-2). When it is determined that there is a problem (in the case of “NO” in step S7-2), the control unit 21 of the customer confirmation support server 20 executes a warning flag recording process (step S7-3). As a result, it is possible to alert a transfer destination that has had a problem in the past.

  (13) In the transfer destination verification process of the present embodiment, the control unit 21 of the customer confirmation support server 20 executes a process for specifying the account status of the transfer destination (step S7-4). Thereby, it is possible to call attention to the transfer to an account that may cause a problem depending on the account status.

  (14) In the present embodiment, the control unit 21 of the customer confirmation support server 20 performs a comparison process with the attention pattern (steps S4-6, S6-7, S7-5). When it is determined that there is similarity (in the case of “YES” in steps S4-7, S6-8, and S7-6), the control unit 21 of the customer confirmation support server 20 executes a warning flag recording process (step S4-7). S4-8, S6-9, S7-7). As a result, it is possible to alert a transaction common to patterns that may cause problems.

In addition, you may change the said embodiment as follows.
In the above embodiment, the control unit 21 of the customer confirmation support server 20 executes a determination process as to whether or not alerting is necessary (step S1-3). Specifically, the confirmation management unit 212 of the control unit 21 determines that an alert is required when an alert flag is recorded in the transaction request management data 260 as a result of various verification processes. Here, each pattern may be weighted. In this case, weighting values are set for items included in various patterns recorded in each pattern information storage unit. Furthermore, information related to the alert level (reference score) for determining whether or not alert is required is registered in the customer confirmation support server 20. Next, a comprehensive evaluation score is calculated by summing up the weight values for the items that do not match the pattern. Then, the control unit 21 of the customer confirmation support server 20 compares the overall evaluation score with the alert level, and determines whether or not alert is required. Furthermore, the weighting value and the alert level may be changed according to the transaction details and customer attributes. For example, when the transaction amount is high or the age of the customer is high, the attention level is lowered and the attention is actively raised by detecting a mismatch in the pattern matching of the general pattern, customer pattern, and individual pattern. be able to.
Further, this weighting value may be set for a combination of a plurality of items. Thereby, in pattern collation, it can alert about the problem which arises by the combination of a predetermined item.

  In the above embodiment, the control unit 21 of the customer confirmation support server 20 executes a collation result output process (step S1-4). Specifically, the confirmation management unit 212 of the control unit 21 outputs the result of pattern matching to the window terminal 11 and the operator terminal 15. The method of alerting is not limited to this. For example, you may make it output alerting with respect to the supervisor of the store where the automatic teller machine 12 was installed. In this case, information for identifying the monitor terminal held by the monitor at each store is recorded in the customer confirmation support server 20. And the confirmation management part 212 of the control part 21 specifies the shop where the transaction terminal 10 in which the transaction which may produce a problem is performed is installed. Next, the confirmation management unit 212 identifies the monitor terminal held by the monitor in this store. And the confirmation management part 212 outputs alerting with respect to this supervisor terminal. Further, the confirmation management unit 212 may respond to the customer at the counter by outputting a message for guiding to the automatic teller machine 12 to a channel (for example, a counter) different from the transaction channel.

  In the above embodiment, the control unit 21 of the customer confirmation support server 20 performs a hearing process (step S1-6). Specifically, the confirmation management unit 212 of the control unit 21 outputs information necessary for hearing to the window terminal 11 and the operator terminal 15. For example, when there is an inconsistent item in an individual pattern, a general pattern, or a customer pattern, information related to a question for confirming the person is output. When there is a matching item in the attention pattern, information related to a question for confirming the transaction is output. In this case, the information regarding the question may be changed according to the level of the person in charge of the financial institution. In this case, a coping method is registered in the attention pattern information storage unit 25 in correspondence with the person in charge level. And the control part 21 of the customer confirmation assistance server 20 specifies the level of the person in charge using the window terminal 11 and the operator terminal 15, and uses the coping method according to this level.

  In the above embodiment, when it is determined that there is usage history accumulation (in the case of “YES” in step S2-8), the control unit 21 of the customer confirmation support server 20 executes past feature information specifying processing. (Step S2-9). Here, information relating to biometric authentication, gender, age, face image, and gender is used as the feature information. And the control part 21 of the customer confirmation assistance server 20 performs the determination process about whether the characteristics correspond (step S2-10). The feature information is not limited to this. For example, it is also possible to use handwriting written on a transaction application form. In this case, the handwriting is converted into data, and the feature amount of the handwriting is calculated.

  In the above embodiment, the control unit 21 of the customer confirmation support server 20 executes behavior information acquisition processing (step S5-6). Here, as the feature amount of the behavior, the amount of change in motion is calculated for the facial expression of the trader and how to move the body. Next, the control part 21 of the customer confirmation support server 20 performs the identification process of the past behavior information of the account holder (step S5-7). And the control part 21 of the customer confirmation assistance server 20 performs the determination process about whether there exists similarity (step S5-8). The feature amount of the behavior is not limited to these. For example, the sweating state may be specified by an image.

  In the above embodiment, when it is determined that the usage history is not accumulated (“NO” in step S2-8), the control unit 21 of the customer confirmation support server 20 executes the general pattern matching process (step S2). -6). On the other hand, if it is determined that there is usage history accumulation (in the case of “YES” in step S2-8) and it is determined that the features match (in the case of “YES” in step S2-10), the customer confirmation support server 20 The control unit 21 executes the individual pattern matching process after the identity state matching process (step S2-11) (step S2-12). The application method of each pattern is not limited to this. For example, even when there is usage history accumulation, the general pattern matching process may be executed if the usage history accumulation amount is equal to or less than the accumulation amount reference value. Further, when the accumulation amount is equal to or less than the accumulation amount reference value, the general pattern and the individual pattern may be used in combination. In this case, both collation results are output to the displays of the window terminal 11 and the operator terminal 15.

  Even when transaction results are accumulated, the reference value for determining the accumulation amount may be changed according to the usage status of the customer. For example, when the variance of each item value of the transaction pattern is smaller than a predetermined variance reference value, the general pattern is preferentially applied. On the other hand, when the variance of each item value of the transaction pattern is larger than a predetermined variance reference value, the individual pattern is preferentially applied. Thereby, the collation pattern used for collation can be specified according to the dispersion | distribution state of a customer's utilization condition.

  In the above embodiment, when necessary information is acquired from the window terminal 11 or the operator terminal 15 and it is determined that the countermeasure is necessary (in the case of “YES” in step S1-5), the control unit 21 of the customer confirmation support server 20 Performs a hearing process (step S1-6). Here, the coping method is not limited to the hearing. For example, the person in charge may be instructed to observe the behavior of the customer. In the customer management data 230, an observation flag may be further recorded in association with the customer code. Then, if an observation flag is recorded in the customer management data 230 specified based on the customer code included in the transaction request acceptance information, the control unit 21 of the customer confirmation support server 20 An alert is output to the display of the terminal 11 or the operator terminal 15. Accordingly, it is possible to detect a transaction that may cause a problem based on the status of a plurality of transactions.

  DESCRIPTION OF SYMBOLS 10 ... Transaction terminal, 11 ... Window terminal, 12 ... Automatic teller machine, 13 ... User terminal, 15 ... Operator terminal, 20 ... Customer confirmation support server, 21 ... Control part, 211 ... Request information acquisition part, 212 ... Confirmation management 213 ... General verification unit, 214 ... Individual verification unit, 215 ... Status verification unit, 216 ... Transfer destination verification unit, 22 ... Individual pattern information storage unit, 23 ... Customer pattern information storage unit, 24 ... General pattern information storage unit 25 ... Attention pattern information storage unit, 26 ... Transaction request storage unit, 30 ... Host system, 32 ... Customer information storage unit, 33 ... Account information storage unit.

Claims (11)

A unique information storage unit in which a unique pattern for each customer generated based on the customer service usage record is recorded in association with the customer identifier;
A general information storage unit in which a general pattern generated based on a customer service usage record that matches the customer attribute is recorded in association with the customer attribute;
A confirmation information storage unit that records confirmation information corresponding to the comparison result of the pattern comparison;
A customer confirmation support system comprising a service providing channel and a connected control unit,
The control unit is
In the service providing channel, obtain a customer identifier and request information of the customer,
When a unique pattern is registered in the unique information storage unit in association with the customer identifier, a pattern comparison is performed between the request information and the unique pattern, and a comparison result is generated.
If the unique pattern is not registered in association with the customer identifier, the customer attribute of the customer is specified,
Perform a pattern comparison between the general pattern associated with the customer attribute in the general information storage unit and the request information, and generate a comparison result,
A customer confirmation support system, wherein information for confirmation corresponding to a comparison result is extracted from the confirmation information storage unit and output. The unique pattern includes customer feature information,
In the service providing channel, obtain feature quantity information of the customer situation,
The customer confirmation support according to claim 1, wherein the feature amount information of the unique pattern and the feature amount information acquired in the service providing channel are compared to generate a comparison result about the customer situation. system. A customer information storage unit in which related party information is recorded in association with the customer identifier;
When it is determined that the customer is not the customer himself / herself based on the comparison result between the feature amount information of the unique pattern and the feature amount information acquired in the service providing channel, the customer information storage unit acquires the related party information of the customer. 3. The customer confirmation support system according to claim 2, wherein confirmation information is extracted from the confirmation information storage unit based on the related party information and output. The feature amount information in the unique pattern recorded in the unique information storage unit is registered for each service type,
In the unique pattern, identify feature quantity information of a service type corresponding to the request information,
4. The customer confirmation support system according to claim 2, wherein the identified feature amount information is compared with the feature amount information acquired in the service providing channel. 5.   5. The unique pattern is updated based on the service use record when information indicating that there is a problem in the service use record is not acquired. Customer confirmation support system described in 1. A caution information storage unit in which a caution pattern for service provision is recorded;
When the information which shows that there was a problem in the service use record is acquired, the caution pattern is updated based on the service use record. The customer confirmation support system described. When request information about the transfer to the transfer destination is acquired, the usage status information about the transfer destination is acquired,
The usage status information is compared with a caution pattern recorded in the caution information storage unit, and information for confirmation is extracted from the confirmation information storage unit based on the comparison result of the pattern comparison and output. The customer confirmation support system according to claim 6.   The customer confirmation support system according to any one of claims 1 to 7, wherein the confirmation information guides to a channel different from the service providing channel.   The customer confirmation support system according to any one of claims 1 to 8, wherein weighting of elements of pattern comparison is changed according to request information from the customer. A unique information storage unit in which a unique pattern for each customer generated based on the customer service usage record is recorded in association with the customer identifier;
A general information storage unit in which a general pattern generated based on a customer service usage record that matches the customer attribute is recorded in association with the customer attribute;
A confirmation information storage unit that records confirmation information corresponding to the comparison result of the pattern comparison;
A method for supporting customer confirmation using a customer confirmation support system including a service providing channel and a control unit connected thereto,
The control unit is
In the service providing channel, obtain a customer identifier and request information of the customer,
When a unique pattern is registered in the unique information storage unit in association with the customer identifier, a pattern comparison is performed between the request information and the unique pattern, and a comparison result is generated.
If the unique pattern is not registered in association with the customer identifier, the customer attribute of the customer is specified,
Perform a pattern comparison between the general pattern associated with the customer attribute in the general information storage unit and the request information, and generate a comparison result,
A customer confirmation support method, wherein information for confirmation corresponding to a comparison result is extracted from the confirmation information storage unit and output. A unique information storage unit in which a unique pattern for each customer generated based on the customer service usage record is recorded in association with the customer identifier;
A general information storage unit in which a general pattern generated based on a customer service usage record that matches the customer attribute is recorded in association with the customer attribute;
A confirmation information storage unit that records confirmation information corresponding to the comparison result of the pattern comparison;
A program for supporting customer confirmation using a customer confirmation support system including a service providing channel and a control unit connected thereto,
The control unit
In the service providing channel, obtain a customer identifier and request information of the customer,
When a unique pattern is registered in the unique information storage unit in association with the customer identifier, a pattern comparison is performed between the request information and the unique pattern, and a comparison result is generated.
If the unique pattern is not registered in association with the customer identifier, the customer attribute of the customer is specified,
Perform a pattern comparison between the general pattern associated with the customer attribute in the general information storage unit and the request information, and generate a comparison result,
A customer confirmation support program that functions as means for extracting and outputting confirmation information corresponding to a comparison result from the confirmation information storage unit.