JP2022175389A - Matching system and matching method - Google Patents

Abstract

To provide a matching system and a matching method capable of selecting a recruiter suitable for customer's characteristics.SOLUTION: A matching system 20 includes: a response control unit 21 for receiving input information entered at a customer terminal via a communication network; a recruiter evaluation database 22 for storing evaluation information about a plurality of pre-registered recruiters; and a matching unit 23. The matching unit 23 estimates customer characteristics from the input information and searches the recruiter evaluation database 22 based on the estimated characteristics to find a recruiter suitable for the customer characteristics. The response control unit 21 transmits information about the found recruiter to the customer terminal via the communication network.SELECTED DRAWING: Figure 1

Description

The present disclosure relates to matching technology for matching between customers and recruiters.

2. Description of the Related Art Conventionally, in various industries, services have been provided for non-face-to-face communication with customers using online call systems such as call center (contact center) systems and web conference systems. In addition, the development of information processing technology for improving services based on voice analysis of customer call data acquired by online call systems is also being actively carried out. Such information processing technology is disclosed, for example, in Japanese Patent Application Laid-Open No. 2021-56612 (Patent Document 1). Patent Literature 1 discloses a call center system that determines a customer's emotion based on the contents of a call and voice and selects an appropriate operator based on the determination result when dealing with a customer in a call center.

JP 2021-56612 A

For example, in the insurance industry, a customer consults with a call center operator about insurance through an online call system, and the operator assigns a solicitor with specialized knowledge of insurance products to the customer based on the content of the consultation. I have something to introduce. However, operators often assign customers to agents who are automatically or simply selected from a large number of agents registered in the system. For this reason, the operator is not always able to introduce agents who are suited to the characteristics of each individual customer, and as a result, the closing rate (the percentage of customers who are introduced by agents who reach a contract) decreases. There is a problem of not improving.

In view of the above, it is an object of the present disclosure to provide a matching system and matching method that enable selection and assignment of recruiters suited to customer characteristics.

A matching system according to a first aspect of the present disclosure is a matching system that communicates with a customer terminal, which is a communication device used by a customer, via a communication network, wherein text information and voice information input at the customer terminal are a response control unit that receives input information from at least one of them via the communication network; a solicitor evaluation database that stores evaluation information about a plurality of pre-registered solicitors; a matching unit for estimating characteristics and searching the sales person evaluation database using the estimated characteristics to find one or more sales representatives suitable for the characteristics of the customer from among the plurality of sales agents; , wherein the response control unit transmits information on the solicitor found by the matching unit to the customer terminal via the communication network.

A matching method according to a second aspect of the present disclosure is a matching method executed in a matching system that communicates with a customer terminal, which is a communication device used by a customer, via a communication network, wherein a step of receiving input information of at least one of text information and voice information via the communication network; a step of estimating characteristics of the customer from the input information; and evaluation information about a plurality of sales agents registered in advance. and searching the salesperson evaluation database using the estimated characteristics to obtain a single salesperson suitable for the characteristics of the customer from among the plurality of salespersons or searching for a plurality of solicitors; and transmitting information about the solicited solicitors to the customer terminal via the communication network.

According to the first and second aspects of the present disclosure, by searching the recruiter database using the customer's estimated characteristics, the customer who is suitable for the characteristics of the customer from among a plurality of registered recruiters Recruiters can be found. Information about the found recruiter is transmitted to the customer terminal. In this way, it is possible to select and allocate a solicitor suitable for the characteristics of each individual customer, so an improvement in the closing rate can be expected.

1 is a block diagram showing a schematic configuration of a matching system according to an embodiment of the present disclosure; FIG. 4 is a flow chart illustrating a processing procedure of a matching method according to an embodiment of the present disclosure;

Next, an embodiment according to the present disclosure and modifications thereof will be described in detail with reference to the drawings. It should be noted that constituent elements denoted by the same reference numerals throughout the drawings have the same configuration and the same function.

FIG. 1 is a block diagram showing a schematic configuration of a matching system 20 according to one embodiment of the present disclosure. As shown in FIG. 1, the matching system 20 is connected to the communication network NW so as to be able to communicate with the customer terminal 10, which is the communication equipment used by the customer 11. FIG. The communication network NW is, for example, SIP (Session Initiation Protocol) or H.264. An IP (Internet Protocol) communication network using a communication protocol such as H.323 can be used, but is not limited to this. The customer terminal 10 may be a mobile communication device such as a mobile phone terminal, or a communication device such as a PC (Personal Computer) that performs fixed communication.

The matching system 20 is configured as an online calling system such as a call center (contact center) system or a web conference system so that voice communication can be performed with the customer terminal 10 via the communication network NW. Referring to FIG. 1, the matching system 20 includes a response control section 21 having a response function of the online call system. This response control unit 21 may be configured to realize an automatic response system that automatically responds to inquiries in natural language of voice information or text information from the customer terminal 10, or alternatively, the customer terminal 10 A human operator (not shown) may be configured to implement a manned response system in which a human operator (not shown) responds to a natural language query of voice or text information from a customer.

The matching system 20 also includes a solicitor evaluation database 22 that stores evaluation information about a plurality of pre-registered solicitors, a matching unit 23 that performs matching between the customer 11 and the solicitor, and an input voice. It is provided with a speech recognition unit 24 for performing speech recognition processing on information AS, and a language processing unit 25 for performing natural language processing (NLP) on input natural language information. The solicitor is assumed to be an insurance solicitor with specialized knowledge of insurance products, but is not limited to this.

The recruiter evaluation database 22, matching unit 23, response control unit 21, speech recognition unit 24, and language processing unit 25 described above may be implemented in a single computer, or may be implemented via a communication channel such as a communication network. It may also be implemented as a distributed processing system by means of multiple computers interconnected via a network.

In the solicitor evaluation database 22, evaluation information indicating the correspondence relationship between one or more persona items representing characteristics of virtual customers and evaluation values (score values) respectively representing evaluations of a plurality of solicitors. is stored in advance. Here, "persona" refers to a virtual persona created to represent a typical user (including customers and agents). The evaluation information may be formed in the form of a lookup table, or may be formed in the form of a neural network model (artificial neural network). For example, user characteristics include name, age, gender, occupation, income, educational background, place of residence, personality, hobbies, preferences, family composition, requests, and disease history. At least one should be used. For example, if the place of residence and age of the customer and the place of residence and age of the solicitor are close to each other, the evaluation value of the solicitor corresponding to the place of residence and age can be increased.

When text information (natural language information) TD transmitted from the customer terminal 10 and received by the response control unit 21 is input, the matching unit 23 estimates the characteristics of the customer 11 from the text information TD, By searching the solicitor evaluation database 22 using the characteristics, one or more solicitors suitable for the characteristics of the customer 11 can be found out of the plurality of registered solicitors. Specifically, the matching unit 23 can search for a recruiter who has a relatively high evaluation value for the estimated characteristics from among a plurality of recruiters registered in the recruiter evaluation database 22 . The response control unit 21 can transmit the solicitor information AD regarding the found solicitor to the customer terminal 10 via the communication network NW. Here, the solicitor information AD includes, for example, the characteristics of the solicitor (gender, face photo, etc.), a self-introduction, and free time.

On the other hand, the voice recognition unit 24 receives voice information AS of the customer 11 (a digital signal representing a voice waveform detected by the microphone of the customer terminal 10) transmitted from the customer terminal 10 and received by the response control unit 21. Then, speech recognition processing such as morphological analysis using an acoustic model and a language model is performed on the speech information AS to generate text recognition information RD. The language processing unit 25 performs natural language processing on the recognition information RD to generate linguistic information LD including uttered text information of the customer 11 and information representing the psychological state of the customer 11 (for example, tone of voice). can do.

When the message information (text information) MD of the customer 11 transmitted from the customer terminal 10 and received by the response control unit 21 is input, the language processing unit 25 performs natural language processing on the message information MD. As a result, the linguistic information LD including text information and information for estimating the mental state of the customer 11 can also be generated.

When the language information LD is input from the language processing unit 25, the matching unit 23 estimates the characteristics of the customer 11 from the language information LD, and searches the solicitor evaluation database 22 using the estimated characteristics. It is possible to search for one or more solicitors suitable for the characteristics of the customer 11 from among the plurality of registered solicitors. The response control unit 21 can transmit the solicitor information AD regarding the found solicitor to the customer terminal 10 via the communication network NW.

Next, an operation example of the matching system 20 described above will be described with reference to FIG. FIG. 2 is a flow chart that schematically illustrates a processing procedure of a matching method according to one embodiment. After the customer 11 operates the customer terminal 10 to establish communication between the customer terminal 10 and the matching system 20, the process of FIG. 2 is started.

Referring to FIG. 2, the response control unit 21 waits until receiving some information from the customer terminal 10 (NO in step S10). When the customer 11 inputs the interview conditions with the solicitor to the customer terminal 10 as the text information TD, the response control unit 21 of the matching system 20 receives the text information TD indicating the interview conditions from the customer terminal 10 via the communication network NW, the message At least one of the information MD and the audio information AS is received (YES in step S10). Interview conditions include, for example, the characteristics of the customer 11 such as gender, desired date and time of interview, desired items (for example, face-to-face consultation or non-face-to-face consultation), and reason for consultation (for example, life events and concerns).

If the response control unit 21 receives only the text information TD and does not receive the message information MD and the voice information AS (NO in step S11 and NO in step S13), the process proceeds to step S21. When the message information MD is received by the response control unit 21 (YES in step S11), the language processing unit 25 performs natural language processing on the message information MD to obtain the text information and the psychological state of the customer 11. Language information LD including information for estimating is generated (step S12). When the response control unit 21 receives the voice information AS (YES in step S13), the voice recognition unit 24 performs voice recognition processing such as morphological analysis on the voice information AS to obtain text recognition information RD. is generated (step S14). Next, the language processing unit 25 performs natural language processing on the recognition information RD to generate the language information LD including the uttered text information of the customer 11 and the information representing the psychological state of the customer 11 (step S15). ).

In step S21, the matching unit 23 estimates the characteristics of the customer 11 from at least one of the input language information LD and text information TD. The matching unit 23 can extract (estimate), for example, words and phrases forming part of the input information (language information LD and text information TD) as information indicating the characteristics of the customer 11 .

In the subsequent step S22, the matching unit 23 searches the solicitor evaluation database 22 using the estimated characteristics to find a single or multiple solicitors suitable for the characteristics of the customer 11 from among the plurality of registered solicitors. find recruiters for The response control unit 21 transmits the solicitor information AD regarding the found solicitor to the customer terminal 10 via the communication network NW (step S23).

After that, when the customer 11 executes an operation input for selecting a solicitor based on the solicitor information AD on the customer terminal 10, the response control unit 21 receives a notification to the effect that the solicitor is selected (step S24). YES), the selected recruiter is notified of the interview request (step S25). Specifically, the response control unit 21 can notify the interview request to the solicitor using communication means such as e-mail. On the other hand, if there is no notification to the effect that a solicitor will be selected (NO in step S24), the process ends.

After face-to-face consultation or non-face-to-face consultation is actually conducted between the customer 11 and the recruiter, the customer 11 inputs the evaluation of the recruiter to the feedback processing unit (not shown) of the matching system 20. In addition, the solicitor can input the presence or absence of the contract to the feedback processing unit.

As described above, the matching system 20 of the present disclosure searches the solicitor evaluation database 22 using the estimated characteristics of the customer 11 to determine the characteristics of the customer 11 from among a plurality of registered solicitors. You can find a suitable recruiter. Information about the found solicitor is transmitted to the customer terminal 10 . In this way, the matching system 20 can select and allocate sales representatives suitable for the characteristics of individual customers, thereby improving the contract rate.

Various embodiments and modifications thereof according to the present disclosure have been described above, but the above-described embodiments and modifications are merely examples, and do not limit the scope of the present invention. It should also be understood that changes, additions and improvements may be made to the above embodiments as appropriate without departing from the spirit and scope of the invention. The scope of the present invention should be construed based on the claims, and should be understood to include equivalents thereof.

The matching system and matching method according to the present disclosure can provide solicitor information suitable for the characteristics of individual customers, and are therefore suitable for use in insurance-related business, for example.

NW: communication network, 10: customer terminal, 11: customer, 20: matching system, 21: response control unit, 22: recruiter evaluation database, 23: matching unit, 24: voice recognition unit, 25: language processing unit.

Claims (8)

A matching system that communicates with a customer terminal, which is a communication device used by a customer, via a communication network,
a response control unit that receives input information of at least one of text information and voice information input at the customer terminal via the communication network;
a solicitor evaluation database in which evaluation information about a plurality of pre-registered solicitors is stored;
By estimating the characteristics of the customer from the input information and searching the salesperson evaluation database using the estimated characteristics, one or more salespersons suitable for the characteristics of the customer are selected from among the plurality of salespersons. Equipped with a matching unit that searches for recruiters,
The response control unit transmits information about the solicitor found by the matching unit to the customer terminal via the communication network.
A matching system characterized by: A matching system according to claim 1,
The solicitor evaluation database stores in advance information indicating a correspondence relationship between one or more persona items representing virtual customer characteristics and evaluation values representing the evaluations of the plurality of solicitors,
The matching unit searches the recruiter evaluation database to find recruiters with relatively high evaluation values for the estimated characteristics from among the plurality of recruiters.
A matching system characterized by: 3. The matching system according to claim 1 or 2, wherein the matching unit includes the customer's name, age, sex, occupation, income, educational background, place of residence, personality, hobbies, tastes, family composition, requests and illness. A matching system characterized by estimating at least one selected from a plurality of types of attributes including histories as a characteristic of the customer. A matching system according to any one of claims 1 to 3,
a voice recognition unit that, when the response control unit receives a digital signal representing a voice waveform as the voice information from the customer terminal, performs voice recognition processing on the digital signal;
A language processing unit that generates language information from the result of the speech recognition processing,
The matching system, wherein the matching unit estimates the characteristics of the customer from the linguistic information. A matching method executed in a matching system that communicates with a customer terminal, which is a communication device used by a customer, via a communication network,
a step of receiving input information of at least one of text information and voice information input at the customer terminal via the communication network;
estimating characteristics of the customer from the input information;
a step of referring to a solicitor evaluation database in which pre-registered solicitor evaluation information regarding a plurality of solicitors is stored;
searching the agent evaluation database using the estimated characteristics to find one or more agents suitable for the characteristics of the customer from among the plurality of agents;
and transmitting information about the found recruiter to the customer terminal via the communication network. The matching method according to claim 5,
The solicitor evaluation database stores in advance information indicating a correspondence relationship between one or more persona items representing virtual customer characteristics and evaluation values representing the evaluations of the plurality of solicitors,
In the step of locating the recruiter, a recruiter with a relatively high evaluation value for the inferred characteristics is searched by searching the recruiter evaluation database.
A matching method characterized by: 7. The matching method according to claim 5 or 6, wherein in the step of estimating the characteristics of the customer, the customer's name, age, sex, occupation, income, educational background, place of residence, personality, hobbies, preferences, family A matching method, wherein at least one selected from a group of attributes including configuration, desires, and disease history is estimated as a characteristic of the customer. A matching method according to any one of claims 5 to 7,
a step of subjecting the digital signal to voice recognition processing when a digital signal indicating a voice waveform as the voice information is received from the customer terminal;
generating language information from the results of the speech recognition process;
A matching method, wherein characteristics of the customer are estimated from the linguistic information.

Images