JP6808774B2 - Approach support system, approach support method and approach support program - Google Patents

Description

The present invention relates to an approach support system, an approach support method, and an approach support program for supporting outbound operations.

At the call center, outbound business that calls prospective customers to introduce new products and other sales, and inbound business that responds to customer calls and supports them are performed. In outbound operations, call center operators operate while interacting with prospective customers. Therefore, a technique for supporting outbound operations in a call center is being studied (see, for example, Patent Document 1). In the technology described in this document, the order status for each customer is registered in the database for each order type, and the outbound request made from the sales department via the network is registered in the database. On the computer terminal operated by the operator, information on the order status and outbound request for each customer is searched from the database, and the search results are displayed in a list on the display device. Call requirements are determined according to the order status of each customer, and the determination result is included in the list display as information for answering telephone calls to customers.

Further, in the outbound business, a technique for improving the customer acquisition rate is also being studied (see, for example, Patent Document 2). The outbound business support device described in this document extracts a set of outbound materials corresponding to the received business contents from the database. Then, the themes and customer knowledge level values whose success rate of outbound business is equal to or higher than a predetermined value are extracted from the database. Then, a set of customer IDs corresponding to each extracted theme and customer knowledge level value is extracted from the database, and customer information corresponding to each customer ID is extracted from the database. Then, the extracted outbound material and the customer information correspondence are associated and output, and transmitted to the outbound operator terminal.

Furthermore, in recent years, word-of-mouth sites, blogs, and SNS (Social Networking Service) have become widespread, and in terms of advertising methods from the corporate side, not only one-sided advertisement transmission by TV commercials, but also with consumers through social media. Marketing through dialogue is attracting attention. Consumers have come to place the highest priority on social media reviews from friends as a motivation for purchasing a product. Therefore, in extracting trends from social media, a technique for extracting trends including users of only mass media and potential users of social media is also being studied (see, for example, Patent Document 3). In the techniques described in this document, the trend analyzer analyzes the metadata collected from the mass media to extract mass media trend keywords. Keyword diffusion degree data is calculated from mass media trend keywords and social media data, and trend keywords suitable for the user's business are extracted using the keyword diffusion degree and product master.

Japanese Unexamined Patent Publication No. 2004-304243 Japanese Unexamined Patent Publication No. 2008-84228 Japanese Unexamined Patent Publication No. 2014-81882

In the above-mentioned outbound business, when approaching a customer, efficient outbound business cannot be realized unless the approach destination is extracted and the timing of the approach is appropriate. On the other hand, outgoing messages in social media services may not be sufficient to reveal potential customer needs.

The present invention has been made to solve the above problems, and an object of the present invention is to provide an approach support system, an approach support method, and an approach support program for improving the efficiency of outbound operations.

(1) The approach support system for solving the above problems includes a transaction information storage unit that records a customer's transaction record, a social media site, and a control unit connected to a person in charge terminal. Then, the control unit sends a transmission message sent by each customer at the time when each customer's transaction is performed for a plurality of customers whose transaction results are recorded in the transaction information storage unit, on the social media site. It is provided with a learning unit which is acquired from the above and generates a prediction model consisting of information common to the acquired plurality of outgoing messages for each product type of the transaction. Further, when a outgoing message matching the prediction model is newly acquired from the social media site, the product of the prediction model is directed to the person in charge terminal and the approach destination that is the originator of the transmission message. It is equipped with an approach destination extraction unit that outputs outbound business instructions for the type. This makes it possible to identify the approach destination in outbound business using social media.

(2) In the approach support system, the control unit is further connected to a customer information storage unit that records the customer attributes of the customer, and the learning unit uses the customer attributes recorded in the customer information storage unit. , It is preferable that a prediction model including common information is generated, and the approach destination extraction unit specifies an approach destination by using the customer attributes included in the prediction model. As a result, the approach destination can be specified in consideration of the customer attribute.

(3) In the approach support system, the control unit is further connected to an account information storage unit that records customer account information, and the approach destination extraction unit stores the account information of the approach destination in the account information storage unit. It is preferable to obtain from the above and use the account information to determine the necessity of outbound business to the approach destination. As a result, the approach destination can be specified in consideration of the account information.

(4) In the approach support system, the control unit acquires an outgoing message of the approach destination from the social media site, and determines timing information for performing the outbound business based on the outgoing time zone of the outgoing message. Is preferable. As a result, outbound business can be performed in consideration of the approach destination situation identified based on the information acquired from social media.

(5) In the approach support system, the control unit acquires the outgoing message of the approach destination from the social media site, and determines the timing information for performing the outbound business based on the transmission location of the outgoing message. Is preferable. As a result, outbound business can be performed in consideration of the approach destination location specified based on the information obtained from social media.

(6) In the approach support system, the control unit acquires the transaction record of the approach destination from the transaction information storage unit, and determines the timing information for performing the outbound business based on the transaction time zone of the transaction record. It is preferable to do so. As a result, outbound business can be performed in consideration of the approach destination situation specified based on the transaction time zone of the transaction record.

According to the present invention, the efficiency of outbound business can be improved.

The explanatory view of the approach support system of embodiment of this invention. The explanatory view of the processing procedure of the learning process of this embodiment. It is explanatory drawing of the processing procedure of this embodiment, (a) is approach destination extraction processing, (b) is explanatory drawing of approach processing. Explanatory drawing of the prediction model of this embodiment. Explanatory drawing of the prediction model of another embodiment. Explanatory drawing of the prediction model of another embodiment.

Hereinafter, an embodiment in which the approach support system, the approach support method, and the approach support program are embodied will be described with reference to FIGS. 1 to 4. In the present embodiment, it is assumed that the approach destination in the outbound business is specified by using the transmission information in the social media and the attribute information and the transaction information of the customer. Here, it is assumed that a social networking service is used as social media.

As shown in FIG. 1, in the present embodiment, the SNS site 10, the support server 20, the host system 30, and the person in charge terminal 40 connected via the network are used.
The SNS site 10 is a server computer (social media site) that provides a social networking service. In this SNS site 10, the outgoing message registered by the user is associated with the user and published together with the outgoing date and time.

The support server 20 is a computer system that supports outbound operations. The support server 20 includes a control unit 21, a prediction model storage unit 22, and a transaction condition information storage unit 23.

The control unit 21 includes a control means such as a CPU (not shown), a memory such as a RAM and a ROM, and performs processes described later (each process such as a learning stage, an information acquisition stage, an approach destination extraction stage, an outbound support stage, etc.). Then, by executing the approach support program, the control unit 21 functions as a learning unit 211, an information acquisition unit 212, an approach destination extraction unit 213, and an outbound support unit 214.

The learning unit 211 executes a process of generating a prediction model for predicting the relationship between the outgoing message registered by the customer on the SNS site 10 and the product type that has made a transaction with the customer. The learning unit 211 holds data regarding the product type to be learned. As the product type to be learned, products handled in outbound business (for example, mortgage loan) are used. Further, the learning unit 211 holds an evaluation target period table. In this evaluation target period table, the evaluation target period (for example, the number of days) for acquiring the outgoing message in the social networking service is recorded for each product type. Further, the learning unit 211 includes a memory for temporarily storing sample information (product type, customer attribute, outgoing message) used for learning.

The information acquisition unit 212 executes a process of acquiring a message sent by the user from the SNS site 10.
The approach destination extraction unit 213 executes a process of identifying the approach destination based on the outgoing message at the SNS site 10 and the information recorded in the host system 30.
The outbound support unit 214 executes a process of instructing the person in charge terminal 40 to perform outbound work for the specified approach destination.

The prediction model storage unit 22 records a prediction model (data) for identifying the approach destination of the outbound business. This prediction model is registered when the learning process described later is performed.

In the present embodiment, as shown in FIG. 4, the prediction model 220 is configured to include data for identifying a product type for performing outbound business based on events and customer attributes.
In the event data area, data regarding the customer status (event) for identifying the customer to whom this approach is taken is recorded. This event is specified, for example, by using a keyword included in a message transmitted in a specific user situation. Therefore, keywords are recorded in association with each event. Events include, for example, part-time job determination and enrollment.

In the customer attribute data area, data related to customer attributes for identifying the customer to whom the outbound business is approached is recorded.
In the product type data area, data related to an identifier for identifying a product (product type) that performs outbound business is recorded for the approach destination.

The transaction condition information storage unit 23 records a transaction condition management record relating to conditions for conducting a transaction. This transaction condition management record is recorded when a new product type is registered. Data related to the product type and transaction conditions are recorded in the transaction condition management record.

In the product type data area, data related to an identifier for identifying a product performing outbound business is recorded.
In the transaction condition data area, data regarding conditions when conducting a transaction for this product type is recorded. As the transaction conditions, for example, conditions related to the transaction status (whether or not there is a transaction) for the product and the account balance required for conducting the transaction are used.

The host system 30 is a computer system that manages customer information and transaction information. The host system 30 includes a customer information storage unit 32 and a transaction information storage unit 33.
A customer management record for a customer of a financial institution is recorded in the customer information storage unit 32. Customer management records are recorded when a customer opens an account at a financial institution. In the customer management record, data regarding the customer ID, name, account identifier, contact, address, customer attribute, and access destination are recorded.

In the customer ID data area, data regarding an identifier for identifying each customer of the financial institution is recorded.
Data related to the customer's name is recorded in the name data area.
In the account identifier data area, data related to an identifier (head / branch code, type code, account number, etc.) for identifying the account held by the customer is recorded.

Data related to the customer's contact information (telephone number, etc.) and address are recorded in the contact information and address data areas, respectively.
Data related to the customer attributes (age, gender, occupation, etc.) are recorded in the customer attribute data area.

In the access destination data area, data regarding the acquisition destination for acquiring the transmitted message in the social networking service used by this customer is recorded. In this data area, for example, an account of a social networking service used by a user, a URL in which a message is published, and the like are recorded.

The transaction information storage unit 33 records a transaction management record related to a transaction with a customer and an account management record for managing an account opened by the customer. The transaction information storage unit 33 functions as an account information storage unit that records customer account information.
A transaction management record is recorded when a transaction is made with a customer. Data related to the transaction ID, transaction date and time, product type, customer ID, and transaction content are recorded in the transaction management record.

In the transaction ID data area, data relating to an identifier for identifying each transaction is recorded.
In the transaction date / time data area, data relating to the date and time when this transaction was performed is recorded.

In the product type data area, data regarding an identifier for identifying the product in which the transaction has been made is recorded.
In the customer ID data area, data regarding an identifier for identifying a customer who has made a transaction is recorded.
In the transaction content data area, data regarding the transaction content (for example, transaction quantity and amount) performed by this customer is recorded.

In addition, the account management record is registered when an account is opened at a financial institution. This account management record records data related to an account identifier, an account balance, and a deposit / withdrawal history.

In the account identifier data area, data regarding an identifier for identifying each account is recorded.
Data related to the balance of this account is recorded in the account balance data area.
In the deposit / withdrawal history data area, data regarding the transaction date / time, amount, description, and location of deposits and withdrawals in this account are recorded. In addition, when cash is deposited and withdrawn using an automated teller machine (ATM), information on the installation location of the cash automatic teller machine used is recorded.

The person in charge terminal 40 is a computer terminal used by a person in charge of outbound business. The person in charge terminal 40 includes a control unit, a storage unit, an input unit including a keyboard, a pointing device, and the like, and an output unit including a display.

Next, the processing performed by using the above system will be described with reference to FIGS. 2 and 3. In the present embodiment, the learning process for learning the prediction model for specifying the approach destination, the approach destination extraction process for extracting the approach destination, and the approach process for outbound to the extracted approach destination will be described in this order.

(Learning process)
First, the learning process will be described with reference to FIG. In this process, a plurality of customers registered in the host system 30 are used as learning targets for creating the prediction model. Here, the processing target person is specified in a plurality of learning target people, and the following processing is repeated.

First, the control unit 21 of the support server 20 executes the transaction information acquisition process (step S1-1). Specifically, the learning unit 211 of the control unit 21 extracts the transaction management record of the processing target person (learning target person) from the transaction information storage unit 33 of the host system 30.

Next, the control unit 21 of the support server 20 executes a determination process as to whether or not there is a transaction for the product type to be learned (step S1-2). Specifically, the learning unit 211 of the control unit 21 confirms whether or not the extracted transaction management record includes a record in which the product type to be learned is recorded.

When it is determined that the transaction of the product type to be learned is not included (when “NO” in step S1-2), the process for this process target person (learning target person) is terminated.
When it is determined that the transaction of the product type to be learned is included (when “YES” in step S1-2), the control unit 21 of the support server 20 executes the customer attribute information acquisition process (step S1). -3). Specifically, the learning unit 211 of the control unit 21 extracts the customer management record of the learning target person from the customer information storage unit 32 of the host system 30 and acquires the customer attribute. Then, the learning unit 211 temporarily stores the customer attribute in the memory in association with the product type.

Next, the control unit 21 of the support server 20 executes the acquisition process of the social media transmission information (step S1-4). Specifically, the learning unit 211 of the control unit 21 acquires an access destination from the customer management record of the processing target person (learning target person). Then, the learning unit 211 identifies the SNS site 10 used by the processing target person by using the access destination of the customer management record. Next, the learning unit 211 specifies the evaluation target period of this product type from the evaluation target period table. Next, the learning unit 211 acquires the outgoing message of the evaluation target period from the SNS site 10 based on the date when the transaction to be learned is performed. Then, the learning unit 211 temporarily stores the acquired outgoing message in the memory in association with the product type and the customer attribute.
The above process is repeated until all the learning subjects are completed.

Then, the control unit 21 of the support server 20 executes the prediction model registration process (step S1-5). Specifically, the learning unit 211 of the control unit 21 specifies a common customer attribute for each product type temporarily stored in the memory. Next, the common keywords included in the outgoing message are specified as events for each product type and common customer attributes. Then, the learning unit 211 generates a prediction model 220 including the specified event, common customer attributes, and product types, and records it in the prediction model storage unit 22.

(Approach destination extraction process)
Next, the approach destination extraction process will be described with reference to FIG. 3A. In the present embodiment, as shown in FIG. 4, the product type is specified by using the event specified based on the message acquired from the SNS site 10 and the customer attribute recorded in the customer information storage unit 32. Furthermore, the approach destination for outbound business is specified based on the transaction status.

First, the control unit 21 of the support server 20 executes the acquisition process of the social media transmission information (step S2-1). Specifically, the information acquisition unit 212 of the control unit 21 periodically acquires an access destination from the customer management record recorded in the customer information storage unit 32 of the host system 30. Then, the learning unit 211 accesses each SNS site 10 by using the access destination of the customer management record, and acquires the outgoing message sent by the customer.

Next, the control unit 21 of the support server 20 executes the prediction model application process (step S2-2). Specifically, the approach destination extraction unit 213 of the control unit 21 applies the prediction model 220 recorded in the prediction model storage unit 22 to the acquired outgoing message. Here, the approach destination extraction unit 213 confirms whether or not the acquired outgoing message includes a keyword corresponding to the event of the prediction model 220 recorded in the prediction model storage unit 22. When the event of the prediction model 220 is detected in the outgoing message, the approach destination extraction unit 213 acquires the customer attribute from the customer management record of the customer information storage unit 32 of the host system 30. Then, the approach destination extraction unit 213 matches the customer attribute acquired from the customer information storage unit 32 with the customer attribute of the prediction model 220.

Next, the control unit 21 of the support server 20 executes the product identification process (step S2-3). Specifically, when the approach destination extraction unit 213 of the control unit 21 matches the customer attribute acquired from the customer information storage unit 32 with the customer attribute of the prediction model 220 and determines that the prediction model 220 can be applied. The product type of this prediction model 220 is specified.

Next, the control unit 21 of the support server 20 executes the transaction status specifying process (step S2-4). Specifically, the approach destination extraction unit 213 of the control unit 21 acquires the transaction conditions for this product type from the transaction condition information storage unit 23. Next, the approach destination extraction unit 213 acquires the customer ID, the customer's account management record matching the customer's account identifier, and the transaction management record from the transaction information storage unit 33.

Next, the control unit 21 of the support server 20 executes a determination process as to whether or not the approach is possible (step S2-5). Specifically, the approach destination extraction unit 213 of the control unit 21 compares the transaction conditions with the transaction status specified by using the account management record and the transaction management record. Then, the approach destination extraction unit 213 determines that the approach is possible when the transaction status matches the transaction conditions, and determines that the approach is not possible when the transaction status does not match.

If it is determined that the approach is not possible (in the case of "NO" in step S2-5), the control unit 21 of the support server 20 ends the process for this customer.
On the other hand, when it is determined that the approach is possible (when “YES” in step S2-5), the control unit 21 of the support server 20 executes the approach process (step S2-6). This process will be described below.

(Approach processing)
Next, the approach process will be described with reference to FIG. 3 (b). Here, the time zone for approaching the customer is specified by using the transmission status of the outgoing message on the social media and the transaction information recorded in the host system 30.

First, the control unit 21 of the support server 20 executes the acquisition process of the social media transmission history (step S3-1). Specifically, the outbound support unit 214 of the control unit 21 acquires the transmission history of the outgoing message transmitted during the immediate neighborhood regular period from the SNS site 10 for the customer specified as the approach destination. Then, the outbound support unit 214 specifies the usage time zone of the SNS site 10 based on the transmission history.

Next, the control unit 21 of the support server 20 executes the transaction history acquisition process (step S3-2). Specifically, the outbound support unit 214 of the control unit 21 acquires the usage history of the automated teller machine in the account management record of the transaction information storage unit 33 of the host system 30.

Next, the control unit 21 of the support server 20 executes the identification process of the approach time zone (step S3-3). Specifically, the outbound support unit 214 of the control unit 21 predicts a predetermined user situation (for example, at home or going out) based on a keyword included in the outgoing message of the outgoing call history. Then, the outbound support unit 214 specifies a time zone corresponding to the predicted predetermined user situation (for example, at home). Further, the outbound support unit 214 identifies the time zone of absence at home based on the usage history of the automated teller machine. Then, the outbound support unit 214 identifies the approach time zone excluding the absent time zone from the home time zone based on the transmission history in the social networking service and the usage history of the automated teller machine.

Next, the control unit 21 of the support server 20 executes the outbound instruction processing (step S3-4). Specifically, the outbound support unit 214 of the control unit 21 gives an outbound instruction to the person in charge terminal 40. This outbound instruction includes customer information of the approach destination (sender of the outgoing message), product type specified by using the prediction model 220, and information on the time zone at home. In this case, the person in charge makes a telephone call to the approach destination at home time based on the outbound instruction.

As described above, according to the present embodiment, the following effects can be obtained.
(1) In the present embodiment, the control unit 21 of the support server 20 executes the social media transmission information acquisition process (step S1-4) and the prediction model registration process (step S1-5). This makes it possible to generate information for extracting the approach destination based on the outgoing information on the social media.

(2) In the present embodiment, the control unit 21 of the support server 20 executes the social media transmission information acquisition process (step S2-1) and the prediction model application process (step S2-2). Then, the control unit 21 of the support server 20 executes the product identification process (step S2-3). This makes it possible to identify the product to be proposed to the approach destination in the outbound business.

(3) In the present embodiment, the control unit 21 of the support server 20 executes a transaction status specifying process (step S2-4) and a determination process of whether or not an approach is possible (step S2-5). When it is determined that the approach is possible (when “YES” in step S2-5), the control unit 21 of the support server 20 executes the approach process (step S2-6). As a result, outbound work can be performed in consideration of the situation of the approach destination.

(4) In the present embodiment, the control unit 21 of the support server 20 executes the acquisition process of the social media transmission history (step S3-1). As a result, it is possible to take an approach at an appropriate timing in consideration of the customer's situation based on the information transmitted on social media. Further, the control unit 21 of the support server 20 executes the transaction history acquisition process (step S3-2). As a result, it is possible to identify the customer's situation based on the transaction situation and take an approach at an appropriate timing.

The above embodiment may be changed to the following aspects.
-In the above embodiment, it is assumed that a social networking service is used as social media. Social media is not limited to this. It is applicable as long as it is a medium that can identify the user, and for example, an electronic bulletin board system, a blog, or the like can be used.

-In the above embodiment, the prediction model 220 recorded in the prediction model storage unit 22 is configured to include data for specifying a product type for performing outbound business based on events and customer attributes. Then, the product type is specified by using the event specified based on the outgoing message acquired from the SNS site 10 and the customer attribute recorded in the customer information storage unit 32. Furthermore, the approach destination for outbound business is specified based on the transaction status. The prediction model for identifying the approach destination is not limited to the above configuration.

For example, as shown in FIG. 5, funds are funded based on an event specified by a keyword included in an outgoing message acquired from the SNS site 10 and a transaction status (account balance, transaction record, etc.) from the transaction information storage unit 33. You may try to identify your needs. In this case, the control unit 21 of the support server 20 generates a prediction model 221 of a customer who has a financial need based on the outgoing message of the customer who has some transaction and the transaction status. Then, the control unit 21 of the support server 20 identifies a customer who has this financial need as an approach destination. Further, the control unit 21 of the support server 20 compares the transaction status acquired from the transaction information storage unit 33 with the transaction conditions recorded in the transaction condition information storage unit 23, and identifies a product type that matches the transaction status.

Further, the information specified based on the outgoing message acquired from the SNS site 10 is not limited to the event, and information that can identify the situation of the caller can be used.
As in the prediction model 222 shown in FIG. 6, for example, the living situation may be specified based on the outgoing message. In this case, the living situation is predicted based on the keywords included in the outgoing message (for example, "low income", "high expense", etc.).

-In the above embodiment, the control unit 21 of the support server 20 executes the outbound instruction processing (step S3-4). In this case, the person in charge makes a telephone call to the approach destination at home time based on the outbound instruction. The contact method in outbound business is not limited to telephone calls. For example, a message about outbound operations may be sent via the social networking service used by the approach destination.

Next, the technical idea that can be grasped from the above embodiment and another example will be added below.
[A] A transaction information storage unit that records the customer's transaction record,
It is an approach support system equipped with a social media site and a control unit connected to the person in charge terminal.
The control unit
For a plurality of customers whose transaction results are recorded in the transaction information storage unit, the outgoing message sent by each customer at the time when each customer's transaction is performed is acquired from the social media site.
A learning unit that generates a prediction model consisting of information common to the acquired plurality of outgoing messages for each product type of the transaction.
When a new outgoing message matching the prediction model is newly acquired from the social media site, the product type of the prediction model is applied to the person in charge terminal and the approach destination that is the sender of the outgoing message. An approach support system characterized by having an approach destination extraction unit that outputs instructions for outbound work.
[B] The control unit is further connected to a customer information storage unit that records the customer attributes of the customer.
The learning unit generates a prediction model including common information by using the customer attributes recorded in the customer information storage unit.
The approach support system according to [a], wherein the approach destination extraction unit identifies an approach destination by using the customer attributes included in the prediction model.
[C] The control unit is further connected to an account information storage unit that records customer account information.
The approach destination extraction unit is characterized in that the account information of the approach destination is acquired from the account information storage unit, and the necessity of outbound business to the approach destination is determined by using the account information [a]. ] Or the approach support system according to [b].
[D] The control unit acquires the outgoing message of the approach destination from the social media site, and obtains the outgoing message of the approach destination.
The approach support system according to any one of [a] to [c], wherein the timing information for performing the outbound business is determined based on the transmission time zone of the outgoing message.
[E] The control unit acquires the outgoing message of the approach destination from the social media site, and obtains the outgoing message of the approach destination.
The approach support system according to any one of [a] to [d], wherein the timing information for performing the outbound business is determined based on the transmission location of the outgoing message.
[F] The control unit acquires the transaction record of the approach destination from the transaction information storage unit.
The approach support system according to any one of claims [a] to [e], characterized in that timing information for performing the outbound business is determined based on the transaction time zone of the transaction record.
[G] A transaction information storage unit that records customer transaction records,
It is a method for providing approach support using an approach support system equipped with a social media site and a control unit connected to the person in charge terminal.
The control unit
For a plurality of customers whose transaction results are recorded in the transaction information storage unit, the outgoing message sent by each customer at the time when each customer's transaction is performed is acquired from the social media site.
For each product type of the transaction, a learning stage for generating a prediction model consisting of information common to the acquired plurality of outgoing messages, and a learning stage.
When a new outgoing message matching the prediction model is newly acquired from the social media site, the product type of the prediction model is applied to the person in charge terminal and the approach destination that is the sender of the outgoing message. An approach support method characterized by performing an approach destination extraction stage that outputs instructions for outbound work.
[H] A transaction information storage unit that records the customer's transaction record,
It is a program for approach support using an approach support system equipped with a social media site and a control unit connected to the person in charge terminal.
The control unit
For a plurality of customers whose transaction results are recorded in the transaction information storage unit, the outgoing message sent by each customer at the time when each customer's transaction is performed is acquired from the social media site.
A learning unit that generates a prediction model consisting of information common to the acquired plurality of outgoing messages for each product type of the transaction.
When a new outgoing message matching the prediction model is newly acquired from the social media site, the product type of the prediction model is applied to the person in charge terminal and the approach destination that is the sender of the outgoing message. An approach support program characterized by functioning as an approach destination extraction unit that outputs instructions for outbound work.

10 ... SNS site, 20 ... support server, 21 ... control unit, 211 ... learning unit, 212 ... information acquisition unit, 213 ... approach destination extraction unit, 214 ... outbound support unit, 22 ... prediction model storage unit, 23 ... transaction conditions Information storage unit, 30 ... host system, 32 ... customer information storage unit, 33 ... transaction information storage unit, 40 ... person in charge terminal.

Claims (8)

A transaction information storage unit that records customer transaction results,
An approach support system equipped with a control unit connected to a media site that can identify the originator of a message .
The control unit
With respect to a plurality of customers whose transaction results are recorded in the transaction information storage unit, the outgoing message sent by each customer is acquired from the media site at the time when the transaction of each customer is performed.
A learning unit that generates a prediction model consisting of information common to the acquired plurality of outgoing messages for each product type of the transaction.
When a new outgoing message matching the prediction model is newly acquired from the media site, a support unit is provided to specify the product type used for outbound business by the prediction model, with the sender of the outgoing message as the approach destination. An approach support system characterized by the fact that. The control unit is further connected to a customer information storage unit that records the customer attributes of the customer.
The learning unit generates a prediction model including common information by using the customer attributes recorded in the customer information storage unit.
The approach support system according to claim 1, wherein the support unit specifies an approach destination by using the customer attribute included in the prediction model. The control unit is further connected to an account information storage unit that records customer account information.
Claim 1 is characterized in that the support unit acquires the account information of the approach destination from the account information storage unit, and uses the account information to determine whether or not it is necessary to output a message to the approach destination. Or the approach support system according to 2. Wherein the control unit acquires the approach destination outgoing messages from previous texture media site,
The approach support system according to any one of claims 1 to 3, wherein the timing information for outputting the message to the approach destination is determined based on the transmission time zone of the outgoing message. Wherein the control unit acquires the approach destination outgoing messages from previous texture media site,
The approach support system according to any one of claims 1 to 4, wherein the timing information for outputting the message to the approach destination is determined based on the transmission location of the outgoing message. The control unit acquires the transaction record of the approach destination from the transaction information storage unit, and obtains the transaction record of the approach destination.
The approach support system according to any one of claims 1 to 5, wherein timing information for outputting a message to the approach destination is determined based on the transaction time zone of the transaction record. A transaction information storage unit that records customer transaction results,
It is a method for providing approach support using an approach support system equipped with a control unit connected to a media site that can identify the sender of a message .
The control unit
With respect to a plurality of customers whose transaction results are recorded in the transaction information storage unit, the outgoing message sent by each customer is acquired from the media site at the time when the transaction of each customer is performed.
For each product type of the transaction, a learning stage for generating a prediction model consisting of information common to the acquired plurality of outgoing messages, and a learning stage.
When a new outgoing message matching the prediction model is newly acquired from the media site, the prediction model is used with the sender of the outgoing message as the approach destination.
, An approach support method characterized by performing a support stage for specifying the product type used for outbound business . A transaction information storage unit that records customer transaction results,
It is a program for approach support using an approach support system equipped with a control unit connected to a media site that can identify the sender of a message .
The control unit
With respect to a plurality of customers whose transaction results are recorded in the transaction information storage unit, the outgoing message sent by each customer is acquired from the media site at the time when the transaction of each customer is performed.
A learning unit that generates a prediction model consisting of information common to the acquired plurality of outgoing messages for each product type of the transaction.
When a new outgoing message matching the prediction model is newly acquired from the media site, the caller of the outgoing message is set as the approach destination, and the prediction model is used to function as a support unit for specifying the product type used for outbound business . An approach support program characterized by this.