Processing device, processing program, and processing method

Abstract

To enable estimation of conversion values.SOLUTION: At least one processor is configured to perform processing for receiving action history information over a period up to an inquiry by a user, receiving inquiry information on the inquiry by the user, and estimating values of uncompleted conversions associated with the inquiry by the user on the basis of the history information and the inquiry information.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present disclosure relates to a processing device, a processing program, and a processing method for estimating a conversion value. [Background technology]

[0002] Conventionally, so-called tracking technology has been known that tracks and analyzes the route taken by a user to access a specific web page and the user's browsing history on the web page, etc. For example, Patent Document 1 describes a management system that matches an access log with an incoming call history by collating an inquiry telephone number with a destination telephone number.

[0003] Also, various tracking methods have been used to measure conversions (CVs), which are when a user interacts with an advertisement on a webpage and then leads to a specific outcome for the advertiser, such as the conclusion of a contract. Furthermore, the value of a conversion is estimated and used in new advertising and promotion. For example, Patent Document 2 describes that when a user views an advertisement, a conversion notification is sent to an advertisement support server, and incoming call information is sent from a notification server to the advertisement support server, and these pieces of information are linked and managed. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2017-037421 A [Patent Document 2] Patent Publication No. 2022-6617 Summary of the Invention [Problem to be solved by the invention]

[0005] However, a conversion is measured only when an advertisement leads to a specific result, such as the conclusion of a contract or the purchase of a product, and therefore measurement of the conversion and estimation of its value have not been performed if the specific result occurs first.

[0006] The present disclosure has been made in light of the above-mentioned background, and provides a processing device, a processing program, and a processing method capable of estimating a conversion value. [Means for solving the problem]

[0007] According to one aspect of the present disclosure, there is provided "a processing device having at least one processor, the at least one processor being configured to execute processing to receive historical information of a user's actions leading up to making an inquiry, receive inquiry information related to the user's inquiry, and estimate a value of an uncompleted conversion related to the user's inquiry based on the historical information and the inquiry information."

[0008] According to one aspect of the present disclosure, there is provided a "processing program for causing a computer having at least one processor to function as follows: receive historical information of a user's actions up to reaching a specified web page, receive inquiry information related to the user's inquiry via the web page, and estimate a value of an uncompleted conversion related to the user's inquiry based on the historical information and the inquiry information."

[0009] According to one aspect of the present disclosure, there is provided a "processing method executed by at least one processor in a computer having at least one processor, the processing method including the steps of receiving historical information of a user's actions up to reaching a specified web page, receiving inquiry information related to the user's inquiry via the web page, and estimating a value of an uncompleted conversion related to the user's inquiry based on the historical information and the inquiry information." Effect of the Invention

[0010] According to the present disclosure, it is possible to provide a processing device, a processing program, and a processing method capable of estimating the value of a conversion.

[0011] It should be noted that the above effects are merely illustrative for the convenience of explanation and are not limiting. In addition to or instead of the above effects, any effect described in this disclosure or any effect obvious to a person skilled in the art may be achieved. [Brief description of the drawings]

[0012] [Figure 1] FIG. 1 is a diagram showing an outline of the estimation of a conversion value by a processing system according to this embodiment. [Diagram 2] FIG. 2 is a conceptual diagram illustrating a schematic configuration of a processing system 1 according to this embodiment. [Diagram 3] FIG. 3 is a block diagram showing an example of the configuration of the terminal device 100 according to this embodiment. [Figure 4] FIG. 4 is a block diagram showing an example of the configuration of the server device 200 according to this embodiment. [Diagram 5] FIG. 5 is a block diagram showing an example of the configuration of the management device 300 according to this embodiment. [Figure 6A] FIG. 6A is a diagram conceptually showing a tracking table stored in memory 313 of management device 300 according to this embodiment. [Figure 6B] FIG. 6B is a diagram conceptually showing a user information table stored in memory 313 of management device 300 according to this embodiment. [Figure 7A] FIG. 7A is a diagram showing a processing sequence executed among the operator terminal device 100-1, the user terminal device 100-2, the server device 200, and the management device 300 according to this embodiment. [Figure 7B]FIG. 7B is a diagram showing a processing sequence executed among the operator terminal device 100-1, the user terminal device 100-2, the server device 200, and the management device 300 according to this embodiment. [Figure 7C] FIG. 7C is a diagram showing a processing sequence executed among the operator terminal device 100-1, the user terminal device 100-2, the server device 200, and the management device 300 according to this embodiment. [Figure 8A] FIG. 8A is a diagram showing a process flow executed in the management device 300 according to this embodiment. [Figure 8B] FIG. 8B is a diagram showing a process flow executed in the management device 300 according to this embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] The embodiments of the present disclosure will be described with reference to the accompanying drawings, in which the same reference numerals are used to designate the same components.

[0014] 1. Overview of the processing system according to this embodiment As an example, the processing system according to this embodiment is used to estimate the value of a conversion when it is assumed that a user operates an advertisement on a webpage, which then results in a specific outcome (conversion: CV) for the advertiser, such as the conclusion of a contract. Specifically, this processing system estimates the value of the conversion based on history information, such as how the user reached a specific webpage, and call record information (inquiry information) when a call is made by operating an advertisement on the webpage. In particular, this processing system estimates the value of an incomplete conversion (future conversion value), rather than a completed conversion. In other words, this processing system estimates the value of a user's inquiry (click value) and provides the result.

[0015] 1 is a diagram showing an outline of the estimation of conversion value by the processing system according to the present embodiment. Specifically, the conversion value is estimated after a user who has accessed a web page talks to an operator based on call information, and the resultant estimated conversion value information (hereinafter also referred to as estimated conversion value information) is transmitted to a server device.

[0016] The terminal device included in this processing system is a terminal device owned by a user. The terminal device, for example, searches a so-called search engine by the user's operation and accesses a web page. When the user shows interest in an advertisement for a service or the like in the web page, the user calls an operator related to the advertisement using call information displayed in the web page. The operator can propose the most suitable service or the like to the user by calling the user while referring to history information such as, for example, what kind of search the user performed to access the web page before making the call and what information the user obtained before accessing the web page. Therefore, in this processing system, history information is transmitted from the user's terminal device to the management device, and the management device transmits the history information to the operator and transfers the call.

[0017] When the call between the user and the operator is completed, the operator transmits call record information to the management device. Here, the call record information is transmitted from an operator terminal device used by the operator. The management device then estimates the value of incomplete conversions based on the history information and the call record information. The management device then transmits information on the estimated conversion value (hereinafter also referred to as estimated conversion value) to a server device. The server device then treats the received estimated conversion value as a conversion value and performs various analyses related to the user's conversion, for example, by machine learning.

[0018] Here, in order to analyze conversions, a web business that operates a server device may request a business that uses a web page to perform tracking to submit conversion information or conversion value information within a specified period of time. This is because there is a risk of accuracy decreasing if machine learning is not performed within the specified period after a user performs a specified action such as clicking on a web page. For example, the specified period may be a relatively short period such as one or two weeks after a user completes an inquiry. During such a period, conversion may not be completed. However, by using the processing device of the present disclosure, it is possible to provide the estimated CV value as value information of a completed conversion, so that various analyses using machine learning can be performed with high accuracy even in the server device.

[0019] In this disclosure, the term "processing device" simply means a device that performs at least one of the processes related to the present disclosure. In particular, in this embodiment, the "processing device" is described as a management device, but it may be a device other than the management device, or may be a device in which another server device or terminal device is added to the management device.

[0020] In addition, in the present disclosure, the term "user" simply means a person who uses or allows others to use a service provided by the processes of the present disclosure. In other words, a "user" may be any person involved in at least one of the processes of the present disclosure.

[0021] In addition, in this disclosure, the term "web page" simply means information output from a device accessed through communication. In particular, in this disclosure, the term "web page" is described as a web page accessed by searching with a search engine, but it may also be a so-called intranet that is made available for viewing only to specific individuals in advance. Specific examples of web pages include so-called landing pages that are accessed by searching with a search engine or via links in displayed advertisements. Such landing pages typically include introduction information for introducing a specific service or product, an application page for the service or product or a link for accessing the application page, call information, etc. It is preferable that the purpose of the landing page is to increase conversions such as purchases, inquiries, and requests for information on products and services by users who have viewed the page.

[0022] In addition, in this disclosure, "call information" simply means information for enabling a call between users. In particular, in this disclosure, "call information" includes "contact information" used to identify a device to be connected so as to be able to transmit information in order to make a call. In addition, a telephone number is given as an example of "contact information", but it may also be information for identifying a device to be connected in order to make a call using a data communication network. In addition, specific examples of call information include telephone numbers, accounts such as SNS, email addresses, links for chat conversations, and links for so-called online conferences. In addition, the call information may be operated by a user to start a call application different from a browser application program for displaying a web page (hereinafter simply referred to as an "application") to start a call.

[0023] In addition, in the present disclosure, "call record information" simply means information on a record of conversation between a user and an operator. In particular, in the present embodiment, "call record information" includes information related to the content of an inquiry made by a user to an operator by telephone or through a conversation in an online conference, and the content of a response made by an operator to a user by telephone or through a conversation in an online conference. In addition, in the present disclosure, when estimating the value of an incomplete conversion and when re-learning a trained estimation model, for example, chat information, email information, information on a message board for questions and answers, or other information such as inquiry information may be used instead of call record information.

[0024] In addition, in the present disclosure, "history information" simply means information such as a user's behavior history on the web until the user makes an inquiry to an operator. In particular, in this embodiment, "history information" includes information obtained by a user using a search engine or the like until the user reaches a specific web page, information (keywords) used to obtain the information, a route to reach the web page, and the like. Furthermore, "history information" may include information obtained on a specific web page before making a call, information used to obtain the information, and the like. In particular, this information is referred to as user history information. Furthermore, it includes a telephone number assigned as contact information described later, terminal information of the user terminal device 100-2 (identification information such as an identification number, model, and OS), and the like. These are information related to user basic information described later that is different from user history information.

[0025] In addition, in the present disclosure, "user basic information" merely means information related to a specific user who has made an inquiry. In particular, in this embodiment, "user basic information" may include information specific to the specific user (user specific information) and attribute information (user attribute information) linked according to the attributes of the user. "User specific information" may include, for example, age, sex, address, occupation, the calling number of the owned terminal device, the IP address of the owned terminal device, and the like. Furthermore, "user specific information" may include sales information related to inquiries made via a web page. Furthermore, "user attribute information" may include information related to the user's hobbies, tastes, or interests according to, for example, age, sex, or occupation. Furthermore, "user attribute information" may include, for example, information on the area corresponding to the user's address, other related information corresponding to the user's occupation, or other basic information.

[0026] Furthermore, in this disclosure, the names given to each device are merely used to distinguish each device from others, and each device may be given a different name depending on its function.

[0027] In addition, even if the disclosure includes descriptions such as "first" and "second," this does not mean that the description is limited to only the two elements to which the description is attached. Naturally, the disclosure may include a "third," a "fourth," and more elements.

[0028] 2. Configuration of Processing System 1 Fig. 2 is a conceptual diagram showing a schematic configuration of a processing system 1 according to this embodiment. As shown in Fig. 2, the processing system 1 includes a user terminal device 100-2 owned by a user, and a server device 200 that operates a web page. As a result, the user terminal device 100-2 can output a web page based on information received from the server device 200 and allow the user to view it. In the following, when there is no need to distinguish between the terminal devices, they may be referred to as the terminal device 100.

[0029] The processing system 1 also includes an operator terminal device 100-1 owned by an operator, and a management device 300 that identifies access to a web page by a user. This allows the processing system 1 to receive a call from a user terminal device 100-2 via the operator terminal device 100-1.

[0030] The above-mentioned devices in the processing system 1 are connected to each other so as to be able to communicate in both directions via a network 400. The network 400 is, for example, the Internet. That is, the terminal device 100 and the server device 200 are connected to an Internet line by wire or wirelessly, and are capable of transmitting and receiving various information and the like. Note that the network 400 may be a network different from the Internet, or may be an intranet.

[0031] In the example of FIG. 2, only one operator terminal device 100-1 and one user terminal device 100-2 are shown, but multiple operator terminal devices 100-1 and user terminal devices 100-2 that can access a web page may be connected to each other so as to be able to communicate with each other via the network 400.

[0032] Furthermore, although the server device 200 and the management device 300 are described as being single devices, the components and processes of the server device 200 and the management device 300 may be distributed across multiple server devices or cloud server devices.

[0033] Furthermore, although the management device 300 is described as an example of the processing device, the processing device is not limited to the management device 300. For example, the processing device may be configured as the operator terminal device 100-1, the server device 200, or the management device 300, or may be a server device equipped with a machine learning function. That is, a server device equipped with a machine learning function may estimate a conversion value by using the machine learning function.

[0034] 3. Configuration of Terminal Device 100 3 is a block diagram showing an example of the configuration of the terminal device 100 according to this embodiment. The terminal device 100 is a collective term for the operator terminal device 100-1 and the user terminal device 100-2. The terminal device 100 does not need to have all of the components shown in FIG. 3, and may have a configuration in which some components are omitted, or may have other components added. The operator terminal device 100-1 and the user terminal device 100-2 may have different configurations, and may not have the same configuration as the terminal device 100.

[0035] The terminal device 100 is typically a terminal device capable of wireless communication, such as a smartphone, but is not limited to such a device. For example, the terminal device may be a feature phone, a personal digital assistant, a PDA, a laptop computer, a desktop computer, a portable game machine, a stationary game machine, or any other device capable of executing the program according to the present disclosure. In addition, the terminal device 100 in the processing system 1 may include multiple terminal devices, but the terminal devices do not need to be the same type and may be different types of terminal devices.

[0036] 3, the terminal device 100 includes an output interface 111, a processor 112, a memory 113 including a RAM, a ROM, or a non-volatile memory (in some cases, an SSD), a communication interface 114 including a communication processing circuit and an antenna, and an input interface 115 including a touch sensor and a hard key. These components are electrically connected to each other via control lines and data lines.

[0037] The output interface 111 functions as an output unit that outputs, in response to an instruction from the processor 112, an image captured by a camera (not shown) included in the input interface 115 and various displays output by executing a program according to the present disclosure to a device such as a display or a printer. In addition, in response to an instruction from the processor 112, the output interface 111 outputs sounds based on acquired information and various sounds output by executing a program according to the present disclosure to a device such as a speaker. Note that such a display is, for example, a liquid crystal display, an organic EL display, or electronic paper. In addition, the output interface 111 includes a speaker and is capable of emitting sounds in response to an instruction from the processor 112.

[0038] The processor 112 is configured with a CPU (microcomputer) and functions as a control unit that controls other components connected thereto based on various programs stored in the memory 113. Specifically, the processor 112 reads out from the memory 113 a program for executing an application according to the present disclosure and a program for executing an OS (Operating System), and executes the program. The processor 112 may be configured with a single CPU, or may be configured with a combination of multiple CPUs and GPUs. In the present disclosure, the processor 112 executes various processes and the like in the sequences of Figs. 7A to 7C in particular.

[0039] The memory 113 includes a main storage device such as a ROM, a RAM, a non-volatile memory, and an auxiliary storage device such as a HDD or an SSD, and functions as a storage unit. The ROM stores instructions for executing the application and the OS according to the present disclosure as a program. The RAM is used to write and read data while the program stored in the ROM is being processed by the processor 112. The non-volatile memory is a memory into which data is written and read by the execution of the program, and the data written therein is saved even after the execution of the program is terminated. In the present disclosure, the memory 113 stores programs for various processes and the like in the sequences of Figures 7A to 7C in particular.

[0040] The communication interface 114 functions as a communication unit that transmits and receives information to and from the remotely installed server device 200, or other terminal devices or server devices, via a communication processing circuit and an antenna. The communication processing circuit processes programs and various information used in the processing system 1 to transmit and receive information to and from other terminal devices or other server devices according to the progress of processing.

[0041] The communication processing circuit processes based on a wideband wireless communication system such as the 5G system, but can also process based on a system related to narrowband wireless communication such as wireless LAN such as IEEE802.11 or Bluetooth (registered trademark) or a system related to non-contact wireless communication. Also, wired communication can be used instead of or in addition to wireless communication.

[0042] The input interface 115 is composed of a touch sensor, hard keys, a camera, a microphone, etc., and functions as an input unit that accepts instruction inputs related to the execution of the program according to the present disclosure and inputs for registering various information. The touch sensor is arranged so as to cover the output interface 111, and transmits information on position coordinates corresponding to image data output from the output interface 111 to the display to the processor 112. As the touch sensor method, known methods such as a resistive film method, a capacitive coupling method, and an ultrasonic surface acoustic wave method can be used. In the present disclosure, the touch sensor detects a swipe operation or a tap operation on each icon displayed on the output interface 111 by an indicator. Note that, although the input interface 115 provided in the terminal device 100 is used in the present disclosure, it is also possible to use an input interface 115 connected wirelessly or by wire to a main body provided with the processor 112, such as a mouse. The camera and microphone included in the input interface 115 are devices capable of detecting external images and sounds. The camera and microphone may be built into the terminal device 100, or may be an external device connected wirelessly or by wire so as to be able to communicate with the terminal device 100.

[0043] 4. Configuration of the Server Device 200 Fig. 4 is a block diagram showing an example of the configuration of the server device 200 according to this embodiment. The server device 200 does not need to have all of the components shown in Fig. 4, and it is possible to adopt a configuration in which some components are omitted, or other components can be added. In addition, the server device 200 does not need to have the components shown in Fig. 4 in a single housing, and it is also possible to distribute the components and processes of the server device 200 to multiple server devices or cloud server devices.

[0044] 4, the server device 200 includes a processor 212 including a CPU or the like, a memory 213 including a RAM, a ROM, a non-volatile memory, an HDD, and the like, and a communication interface 214. These components are electrically connected to each other via control lines and data lines.

[0045] The processor 212 is configured with a CPU (microcomputer) and functions as a control unit for controlling other components connected thereto based on various programs stored in the memory 213. Specifically, the processor 212 reads out from the memory 213 a program for executing an application according to the present disclosure and a program for executing an OS, and executes the program. The processor 212 may be configured with a single CPU, or may be configured with a combination of multiple CPUs and GPUs. In the present disclosure, the processor 212 executes various processes and the like in the sequences of Figs. 7A to 7C in particular.

[0046] The memory 213 includes a RAM, a ROM, a non-volatile memory, and a HDD, and functions as a storage unit. The ROM stores instructions for executing the application and the OS according to the present disclosure as a program. Such a program is loaded and executed by the processor 212. The RAM is used to write and read data while the program stored in the ROM is being processed by the processor 212. The non-volatile memory is a memory into which data is written and read by the execution of the program, and the data written therein is stored even after the execution of the program is terminated. In the present disclosure, the memory 213 stores programs for various processes and the like in the sequences of Figures 7A to 7C in particular.

[0047] The communication interface 214 functions as a communication unit that transmits and receives information to and from the terminal device 100 and other server devices installed remotely via a communication processing circuit and an antenna. The communication processing circuit processes the programs and various information used in the processing system 1 to transmit and receive information from the terminal device 100 and other server devices according to the progress of processing. The communication processing circuit processes based on a wideband wireless communication method such as the 5G method, but can also process based on a method related to narrowband wireless communication such as wireless LAN and Bluetooth (registered trademark) such as IEEE802.11, or a method related to non-contact wireless communication. In addition, wired communication can be used instead of or in addition to wireless communication.

[0048] 5. Configuration of management device 300 Fig. 5 is a block diagram showing an example of the configuration of the management device 300 according to this embodiment. The management device 300 does not need to have all of the components shown in Fig. 5, and it is possible to adopt a configuration in which some components are omitted, or other components can be added. In addition, the management device 300 does not need to have the components shown in Fig. 5 in a single housing, and each component and process of the management device 300 can be distributed to multiple server devices or cloud server devices.

[0049] 5, the management device 300 includes a processor 312 configured with a CPU or the like, a memory 313 including a RAM, a ROM, a non-volatile memory, an HDD, and the like, and a communication interface 314. These components are electrically connected to each other via control lines and data lines.

[0050] The processor 312 is composed of a CPU (microcomputer) and functions as a control unit for controlling other components connected thereto based on various programs stored in the memory 313. Specifically, the processor 312 reads out from the memory 313 a program for executing an application according to the present disclosure and a program for executing an OS, and executes the program. The processor 312 may be composed of a single CPU, or may be composed of a combination of multiple CPUs and GPUs. The processor 312 particularly executes "a process of receiving history information of a user's actions until reaching a specific web page," "a process of receiving inquiry information related to a user's inquiry via a web page," "a process of estimating the value of an incomplete conversion related to a user's inquiry based on the history information and the inquiry information," and the like.

[0051] In addition, the processor 312 performs re-learning of the trained estimation model used to estimate the value of an incomplete conversion. For example, the processor 312 may perform re-learning of the trained estimation model using the user's behavior history, the user's unique information, the user's attribute information linked to the unique information, the call record information, or the value calculated from the completed conversion. For the re-learning, any one of these pieces of information may be used, and it is not necessary that all of them are available.

[0052] The memory 313 includes a RAM, a ROM, a non-volatile memory, and a HDD, and functions as a storage unit. The ROM stores instructions for executing the application and the OS according to the present disclosure as a program. Such a program is loaded and executed by the processor 312. The RAM is used to write and read data while the program stored in the ROM is being processed by the processor 312. The non-volatile memory is a memory into which data is written and read by the execution of the program, and the data written therein is stored even after the execution of the program is terminated. In the present disclosure, the memory 313 stores programs for various processes and the like in the sequences of Figures 7A to 7C in particular.

[0053] In the present disclosure, the memory 313 stores programs for "receiving history information of a user's actions until reaching a specific web page," "receiving inquiry information related to a user's inquiry via a web page," "estimating the value of an incomplete conversion related to a user's inquiry based on the history information and the inquiry information," etc. The memory 313 also stores the above-mentioned trained estimation model, as well as a program for calculating the value of a completed conversion.

[0054] The communication interface 314 functions as a communication unit that transmits and receives information to and from the terminal device 100 and other server devices installed remotely via a communication processing circuit and an antenna. The communication processing circuit processes the programs and various information used in the processing system 1 to transmit and receive information from the terminal device 100 and other server devices according to the progress of processing. The communication processing circuit processes based on a wideband wireless communication method such as the 5G method, but can also process based on a method related to narrowband wireless communication such as wireless LAN and Bluetooth (registered trademark) such as IEEE802.11, or a method related to non-contact wireless communication. In addition, wired communication can be used instead of or in addition to wireless communication.

[0055] 6. Information stored in memory 313 of management device 300 6A is a diagram conceptually showing a tracking table stored in memory 313 of management device 300 according to this embodiment. In the tracking table, an administrator ID, a web ID, a contact ID, a user ID, estimated CV value information, and CV value information are stored in association with a tracking ID. Note that the information stored in these tracking tables is merely an example, and it is not necessary to satisfy all of this information, and information other than this information may be stored in the tracking table.

[0056] The "tracking ID" is an identification number given to each tracking process. The "tracking ID" is a number given when the management device 300 acquires history information such as a search history in the user terminal device 100-2 (hereinafter, also referred to as tracking) and the user makes a call. The tracking process is a process that starts when the user makes a call. The "administrator ID" is an identification number given to each company or the like that is the administrator of a web page. The "web ID" is an identification number given to each web page. Therefore, when one administrator manages one web page, one "web ID" is assigned to the "administrator ID". On the other hand, when one administrator manages multiple web pages, multiple "web IDs" are assigned to the "administrator ID". Also, when multiple administrators jointly manage web pages, multiple "administrator IDs" are assigned to one "web ID".

[0057] The "contact ID" is an identification number given to each piece of contact information (e.g., a telephone number) for monitoring calls made by a user. The contact information is, for example, a telephone number used for tracking. This telephone number is a telephone number owned by the management device 300. When the user terminal device 100-2 calls the telephone number, the operator terminal device 100-1 makes a call via the management device 300. A single or multiple "contact IDs" are assigned to one "web ID". Specifically, when multiple user terminal devices 100-2 access a web page corresponding to the "web ID", the "contact ID" assigns different contact information to each user terminal device 100-2. The "user ID" is an identification number given to each user. Specifically, the "user ID" is an identification number assigned to a user that is inferred from the individual number of the user terminal device 100-2 that accessed the web page. By accumulating information in the tracking table in this way, unique information for each user is accumulated.

[0058] "Estimated conversion value information" is value information of incomplete conversions estimated based on information obtained by tracking. Here, the information obtained by tracking corresponds to the above-mentioned history information and call record information. Here, the "estimated conversion value information" is generated by a trained estimation model described later, and is stored in the memory 313 of the management device 300 each time a conversion value is estimated.

[0059] The "CV value information" is value information of a completed conversion that is determined by information obtained by tracking. For example, the "CV value information" is value information that is determined when a user operates an advertisement on a web page and then a specific result, such as a contract being concluded, actually occurs for the advertiser. The "CV value information" is not determined unless a specific result actually occurs, so it is not stored at the time the call is completed because a contract has not been concluded, but is stored in the memory 313 of the management device 300 when a contract is concluded thereafter. The "CV value information" is calculated by the processor 312 executing a program stored in the memory 313. The "CV value information" may also be stored and managed separately in a customer relationship management (CRM) tool of the user. A typical sales management tool is an example of a customer relationship management tool.

[0060] FIG. 6B is a diagram conceptually illustrating a user information table stored in the memory 313 of the management device 300 according to this embodiment. The user information table is an information table formed by accumulating information stored in the tracking table. In the user information table, user history information, user specific information, user attribute information (external data ID), and call record information are stored in association with the user ID. Note that the information stored in these user information tables is merely an example, and it is not necessary to satisfy all of these pieces of information, and information other than these pieces of information may be stored in the user information table. Note that the user basic information of the present disclosure is composed of the user specific information and the user attribute information.

[0061] The "user history information" is information including the history of the user's actions. The user history information is generated based on the information stored in the tracking table and tables corresponding to the information. Specifically, the user history information includes the access history of the web pages accessed by the user using the user terminal device 100-2, the date and time of the call, and the contact information used for the call. The "user specific information" is information specific to the user that can be determined from what is included in the above-mentioned history information. That is, for example, the "user specific information" corresponds to the telephone number assigned as the contact information included in the history information, and the terminal information of the user terminal device 100-2 (identification information such as identification number, model, OS, etc.). Furthermore, the "user specific information" may include, for example, age, sex, address, occupation, the calling number of the owned terminal device, the IP address of the owned terminal device, etc. And, the "user specific information" may include the sales information of the user related to the inquiry via the web page and multiple conversion value information managed by the above-mentioned customer relationship management tool.

[0062] The "user attribute information" is data obtained from the outside, and in other words may be stored as an external data ID. For example, the "user attribute information" may include information on the hobbies of the user, which is accumulated for each user group in which a plurality of users are classified according to a predetermined attribute. Here, the attribute may include at least one of the user's gender, age, or income. The "user attribute information" may include area information corresponding to the user's address, other related information corresponding to the user's occupation, or other basic information. Various attribute information that is the basis of the "user attribute information" may be separately stored in advance in the memory 313 of the management device 300, and the user attribute information may be extracted from the various attribute information by determining user-specific information so as to correspond to the user-specific information. The "call record information" is information that includes the contents of a call between the user and an operator. For example, the "call record information" may include a call history, such as the contents of answers given to an operator when making a call, and a response history to a questionnaire, etc. By forming a user information table in this way, user-specific information can be collectively managed.

[0063] The memory 313 may store a management table combining the above-mentioned tracking table and user information table, instead of storing the tracking table and user information table separately. In this case, the processor 312 of the management device 300 may appropriately generate a table composed of necessary information from the one table. In addition, although the above description has been given with the user information and the external information separated, the user information may be stored in the external information, or the external information may be stored in the user information.

[0064] 7. Processing sequence performed by processing system 1 7A to 7C are diagrams showing a processing sequence executed among the operator terminal device 100-1, the user terminal device 100-2, the server device 200, and the management device 300 according to this embodiment. The flow of the processing and information provided to the user will be described below with reference to each of the drawings.

[0065] (A. Setting Registration) First, FIG. 7A is a diagram showing a processing sequence for registering settings for performing tracking processing in a web page in the processing system 1 according to this embodiment. As shown in FIG. 7A, the management device 300 sets the web page to perform the tracking processing (S11). Specifically, the processor 312 of the management device 300 sets settings for displaying contact information, etc. These settings are configured in a programming language such as JavaScript (registered trademark). After setting (S11), the management device 300 transmits a request for registering the settings to a web page stored in the server device 200 via the communication interface 314 (T11). The server device 200 receives the request via the communication interface 214 and registers the settings. Specifically, the management device 300 assigns a web ID to the web page stored in the server device 200 and stores it in a web table (not shown). In addition, the management device 300 assigns an administrator ID to an administrator who manages the web page, and stores it in an administrator table (not shown).

[0066] (B. Tracking Processing) Next, Fig. 7B and Fig. 7C are diagrams showing a processing sequence of the processing executed on the above-mentioned web page in the processing system 1 according to this embodiment. As shown in Fig. 7B, the user terminal device 100-2 executes a web search process in response to a user's operation (T21). Specifically, in the web search process, a search engine searches for web pages related to keywords arbitrarily set by the user. The user terminal device 100-2 repeats the web search process in response to a user's operation, and when the user selects a link to the above-mentioned web page, the user terminal device 100-2 requests the server device 200 to browse the web page (T22). The browsing request (T22) includes terminal information such as the identification number, model, and OS identification information of the user terminal device 100-2.

[0067] When the server device 200 receives the browsing request (T22), it requests the management device 300 for a phone number for tracking as contact information (T23). The management device 300 stores a plurality of phone numbers for tracking in a contact table (not shown) that is associated with a contact ID, and responds to the server device 200 by assigning at least one of the plurality of phone numbers to the browsing request (T22) (T24). When the server device 200 receives the number response (T24) from the management device 300, it generates display information to be displayed as a web page and acceptance information for accepting an operation by a user (S21). Specifically, the server device 200 generates the display information and acceptance information by specifying the angle of view of the web page, the size and position of the displayed icons, etc., and the programming language for processing executed by the user's operation, etc., based on the terminal information received in the browsing request (T22). Furthermore, the server device 200 transmits (T25) the display information, the acceptance information (S21), and response information including the telephone number to the user terminal device 100-2 that made the viewing request (T22).

[0068] The user terminal device 100-2 outputs a web page on the display in accordance with the viewing response (T25) received from the server device 200 (S22). For example, an icon displaying the above-mentioned telephone number is arranged on the web page. The user performs an operation to call the telephone number, for example, by tapping the icon. That is, the user terminal device 100-2 accepts a call operation from the user when the icon arranged on the web page is tapped (S23).

[0069] Thereafter, the user terminal device 100-2 transmits the history information to the management device 300 (T26). Here, the history information includes link information of the web page, a telephone number as contact information assigned when responding to the browsing request (T22) (T24), terminal information of the user terminal device 100-2, and the like. When the management device 300 receives the history information (T26), it stores the history information in a tracking table and a user information table (S24). Specifically, the management device 300 generates a tracking ID and stores it in the tracking table, stores each ID associated with the tracking ID in the tracking table (FIG. 6A), stores user-specific information included in the history information in user-specific information, and stores the history information (particularly the search history of the web search (T21) and the date and time of access to the web page) in the user history information (FIG. 6B).

[0070] More specifically, the management device 300 identifies a web ID from a web table based on link information of the web page included in the history information, for example. The management device 300 also identifies an administrator ID from an administrator table based on a web ID, for example. The management device 300 also identifies a contact ID from a contact table based on a telephone number as contact information included in the history information, for example. The management device 300 also generates a contact ID based on terminal information of the user terminal device 100-2 included in the history information, for example. The management device 300 then stores these IDs in a tracking table. In addition to generating a contact ID, the management device 300 may also identify a user ID from a user information table based on terminal information of the user terminal device 100-2. In addition, it has been described that the user terminal device 100-2 transmits history information to the management device 300, but the history information may be stored in the server device 200, and the server device 200 may transmit the history information to the management device 300. Note that the management device 300 additionally stores in the memory 313, for example, information related to the user, the user's terminal information, and the web search history included in the history information, if the user-specific information is not stored in the user information table.

[0071] After that, as shown in FIG. 7C, the user terminal device 100-2 calls the above-mentioned telephone number by the user's operation (T30). When the management device 300 receives a call from the user terminal device 100-2 to the telephone number, it identifies the user based on the calling number (i.e., the telephone number as contact information assigned when responding (T24) to the browsing request (T22)) (S31). Specifically, the management device 300 identifies a contact ID from a contact table based on the calling number, and identifies a tracking ID from a tracking table (FIG. 6A). In addition, the management device 300 identifies a user ID corresponding to the tracking ID. Note that, although it has been described above that the user is identified based on the calling number (S31), the user may be identified based on the called number (i.e., the telephone number of the user terminal device 100-2). In this case, the management device 300 may identify user information including the called number from a user information table, and identify the corresponding user ID.

[0072] The management device 300 identifies user specific information corresponding to the identified user ID from the user information table, and extracts user history information (particularly the search history of the web search (T21) and the access date and time of the web page) (S32). After that, the management device 300 transfers the call from the user terminal device 100-2 to the operator terminal device 100-1 (T31). In addition, the management device 300 transmits the history information to the operator terminal device 100-1 (T32). Then, the operator terminal device 100-1 executes a call process with the user terminal device 100-2 while allowing the operator to view the history information (S33). That is, the operator can make a call with the user while viewing the history information. Note that, although it has been described above that the history information is transmitted to the operator terminal device 100-1 (T32), all of the information contained in the user information may be transmitted to the operator terminal device 100-1, or a part of the user information may be transmitted to the operator terminal device 100-1.

[0073] After that, when the call ends, the operator terminal device 100-1 transmits the call record information, which is the conversation between the user and the operator, to the management device 300 via the communication interface 114 (T33). When the management device 300 receives the call record information (T33), it stores the call record information in the user information table (S34). Specifically, the management device 300 specifies the recording destination of the received call record information based on the user ID included in the call record information. Here, the management device 300 can store the received call record information as voice information, but may also convert it to text and store it for estimating the CV value, which will be described later.

[0074] Next, the management device 300 estimates the value of an incomplete conversion related to the user's inquiry to the operator based on the received information (S35). Here, the received information is history information and call record information. The method of estimating the CV value will be explained in detail later. Thereafter, the management device 300 transmits the estimated CV value information, which is the estimation result, to the server device 200 via the communication interface 314 (T34). This makes it possible for the server device 200 to perform various analyses related to the user's conversion based on the received estimated CV value information.

[0075] 8. Processing flow executed in the user terminal device 100-2 The process flow executed in the user terminal device 100-2 will now be described in detail. (A. Estimation of CV value) Fig. 8A is a diagram showing a process flow executed in management device 300 according to this embodiment. Specifically, Fig. 8A shows a process flow executed to estimate a CV value in management device 300. This process flow is mainly executed by processor 312 of management device 300 reading and executing a program stored in memory 313.

[0076] 8A, in step S101, processor 312 of management device 300 determines whether or not history information has been received via communication interface 314. If processor 312 of management device 300 has not received history information, it does not proceed to the next step (S101: No), and if it has received history information, it proceeds to S102 (S101: Yes).

[0077] In step S102, the processor 312 of the management device 300 stores the received history information in the memory 313. For example, the processor 312 stores the history information in a tracking table and a user information table of the memory 313. Specifically, the processor 312 generates a tracking ID and stores it in the tracking table, stores each ID associated with the tracking ID in the tracking table, and stores various information included in the history information in the user history information and user specific information in the user information table.

[0078] Next, in step S103, the processor 312 of the management device 300 identifies the user. Specifically, the processor 312 identifies a contact ID from the contact table based on the calling number, and identifies a tracking ID from the tracking table (FIG. 6A). This enables the processor 312 to identify a user ID corresponding to the tracking ID.

[0079] Next, in step S104, the processor 312 of the management device 300 determines whether or not the call record information has been received via the communication interface 314. If the processor 312 of the management device 300 has not received the call record information, it does not proceed to the next step (S104: No), and if the processor 312 has received the call record information, it proceeds to S105 (S104: Yes).

[0080] In step S105, the processor 312 of the management device 300 stores the received call record information in the memory 313. For example, the management device 300 specifies a recording destination of the received call record information based on a user ID included in the call record information, converts voice data that has not been converted to text, etc., into text, and stores the call record information in the specified recording destination.

[0081] In step S106, the processor 312 of the management device 300 reads three types of information as a preparation step for estimating the CV value. Specifically, the processor 312 reads the user history information, the user specific information, and the call record information recorded in the user information table. Then, in step S107, the processor 312 of the management device 300 inputs these pieces of information into the trained estimation model to estimate the value information of the incomplete CV. Here, since the user history information and the user specific information are included in the history information, in other words, the processor 312 uses the history information and the corresponding call record information. After that, in step S108, the processor 312 of the management device 300 transmits the estimated CV value information, which is the estimation result, to the server device 200 via the communication interface 314.

[0082] (B. Creation and Use of Trained Estimation Models) Fig. 8B is a diagram showing a processing flow executed in the management device 300 according to this embodiment. Specifically, Fig. 8B is a diagram showing a processing flow relating to the generation to use of a trained estimation model according to this embodiment. Note that this processing flow may be executed by the processor 312 of the management device 300, or may be executed by a processor of another processing device.

[0083] According to FIG. 8B, the processor 312 reads out information (history information and call record information) acquired by, for example, T26 in FIG. 7B and T33 in FIG. 7C as learning data (S111). Specifically, the processor 312 reads out user history information and user specific information from the user information table in FIG. 6B stored in association with the history information. The processor 312 also reads out user attribute information linked to the user specific information included in the history information as learning data (S111). Specifically, the processor 312 reads out user attribute information from the user information table in FIG. 6B stored in association with the history information. In other words, the processor 312 uses information before the user's inquiry (history information) and the actual user's inquiry (call record information) to generate and further learn a trained estimation model.

[0084] Furthermore, the processor 312 also reads out completed CV value information as learning data (S112). Specifically, the processor 312 may extract the CV value information corresponding to the history information and call record information read out in S111 from the tracking table in Fig. 6A, or may read it out from the above-mentioned customer relationship management tool. In other words, the processor 312 uses the information (CV value information) obtained after the negotiation is completed as information of the correct answer in the trained estimation model.

[0085] When the above-mentioned history information (behavior history, user specific information), user attribute information, call record information, and CV value information are obtained, the processor 312 executes a step of performing machine learning of an estimation pattern of a CV value using these (S113). As an example, the machine learning is performed by providing a set of these pieces of information to a neural network that combines neurons, and repeating learning while adjusting the parameters of each neuron so that the output from the neural network is the same as the correct information. Then, the processor 312 executes a step of acquiring a trained judgment model and storing it in the memory 313 (S114). Note that the acquired trained estimation model may be stored in another device connected to the management device 300 via a wired or wireless network.

[0086] The processor 312 repeats the additional learning (iterative learning) each time the above-mentioned learning data is acquired, thereby improving the output accuracy of the trained estimation model.

[0087] On the other hand, when new tracking is performed and a CV value is estimated, the processor 312 reads out the information (history information and call record information) acquired by T26 in FIG. 7B and T33 in FIG. 7C as input data (S115). Then, the processor 312 executes the process of estimating the CV value by inputting the input data into the trained estimation model (S116). After that, the processor 312 stores the obtained estimated CV value information in the tracking table of FIG. 6A in a form linked to the tracking ID (S117). As a result, the estimated CV value information is stored in association with a user ID such as the user's phone number.

[0088] The learning data is not limited to the above-mentioned information, and for example, estimated CV value information may also be used as learning data.

[0089] 9. Variations In the above-mentioned embodiment, the explanation is based on the premise that artificial intelligence (AI) that is a trained estimation model is used, but the present invention is not limited to this. For example, an estimation database may be constructed using various data stored in memory 313, and the CV value may be estimated by referring to the database. Even in such a case, the CV value is estimated based on the user's history information and inquiry information (call record information), and since learning is not required, system management is simplified.

[0090] In the above embodiment, the communication between the user and the operator is based on a telephone call, but a conversation using a Web conference application, a chat, or an e-mail may be used as an inquiry means, or these may be used in combination with a telephone. In such a case, the inquiry information is not limited to the call record information, and inquiry information according to the means used is appropriately used.

[0091] In the above-described embodiment, it is assumed that the CV value is estimated only once, but the CV value may be estimated multiple times within the deadline for providing the estimated CV value information requested by the administrator of the server device 200. That is, the management device 300 may execute the estimation process multiple times before transmitting the estimated CV value information and transmit the latest result to the server device 200. For example, the second or subsequent estimation may be performed in response to additional learning of the learned estimation model. This makes it possible to provide more accurate estimated CV value information. In addition, when value information of a completed CV is obtained by completing a contract or the like within the deadline for providing the information, the management device 300 may calculate the value of the completed CV and transmit the CV value information, which is the calculation result, to the server device 200.

[0092] In the above-described embodiment, it is assumed that the user reaches a specific web page and makes an inquiry from there, but this is not limited to the above. For example, a user who has watched a television commercial may call a telephone number displayed in the television commercial. In this case, the user's history information may be acquired by the inquiry in the call and transmitted to the management device 300 together with the call record information. The management device 300 then estimates the CV value using the received history information (user's unique information, information on watching the television commercial) and the call record information.

[0093] In addition, in the case where the CV value is estimated multiple times, the route leading to the first call and the route leading to the second or subsequent calls may be different. For example, if the first call was made via a TV commercial, the CV value via the commercial may be estimated, and if the same user makes a second call via a web page after that, the CV value via the web page may be estimated. In this case, the estimation result of the first CV value may be updated by the estimation result of the second CV value. In such a case, if there is an error in the unique information or attribute information (e.g., income, etc.) of the user at the time of the first inquiry, and the unique information or attribute information (e.g., income, etc.) of the user at the time of the second inquiry becomes more accurate, the information of the estimated CV value also becomes more accurate. For example, if the original user is a 40-year-old man in the affluent class, the estimated CV value will be calculated lower and more erroneously than if the second inquiry user is correctly determined to be a 40-year-old man in the affluent class, but the estimation of the second CV value makes it possible to update to more accurate data.

[0094] The processes and procedures described in this specification can be realized not only by those explicitly described in this disclosure, but also by software, hardware, or a combination of these. Specifically, the processes and procedures described in this specification can be realized by implementing logic corresponding to the processes in a medium such as an integrated circuit, a volatile memory, a non-volatile memory, a magnetic disk, or an optical storage. In addition, the processes and procedures described in this specification can be implemented as computer programs and executed by various computers including terminal devices and server devices.

[0095] Although the processes and procedures described herein are described as being executed by a single device, software, component, or module, such processes or procedures may be executed by multiple devices, multiple software, multiple components, and / or multiple modules. Furthermore, although the various information described herein is described as being stored in a single memory or storage unit, such information may be stored in multiple memories provided in a single device or multiple memories distributed across multiple devices. Furthermore, the software and hardware elements described herein may be realized by integrating them into fewer components or breaking them down into more components. [Explanation of symbols]

[0096] 100 Terminal Equipment 200 Server device 300 Management device (processing device)

Claims

1. Includes at least one processor, The at least one processor is A camera is used to capture images of the subject of the user, and one or more judgment images of the subject are acquired via the camera. The user's medical interview information and attribute information are obtained, The possibility of contracting a predetermined disease is determined based on a trained judgment model for determining the likelihood of contracting a predetermined disease, and at least one of the judgment image, the medical interview information, and the attribute information. Based on at least one of the aforementioned assessment image, the aforementioned medical interview information, and the aforementioned attribute information, the reliability of the likelihood of the disease is determined. A processing device configured to perform processing for the purpose of

2. The aforementioned trained decision model is A first trained judgment model that receives the judgment image as input and determines the first possibility of contracting the predetermined disease, The processing apparatus according to claim 1, comprising a second trained determination model that receives at least one of the user's medical interview information and attribute information and determines a second possibility of the user suffering from the predetermined disease.

3. The at least one processor is Based on the comparison results of the first possibility and the second possibility, the reliability of the possibility of the disease is determined. The apparatus according to claim 2.

4. The at least one processor is Based on at least one of the user's medical interview information and attribute information, and the distribution information of the training data of the second trained judgment model, the confidence level of the possibility of the disease is determined. The apparatus according to claim 2.

5. The at least one processor is The confidence level of the likelihood of the disease is determined based on whether at least one of the user's medical interview information and attribute information lies inside the convex hull generated from the distribution information of the learning data. The apparatus according to claim 4.

6. The at least one processor is The confidence level of the likelihood of the disease is determined based on the distance between at least one of the user's medical interview information and attribute information and the convex hull generated from the distribution information of the learning data. The apparatus according to claim 4.

7. The at least one processor is The reliability of at least one of the user's medical interview information and attribute information is determined based on the distance between at least one of the user's medical interview information and attribute information and the convex hull generated from the distribution information of the training data. The apparatus according to claim 4.

8. The at least one processor is The reliability of the possibility of the disease is determined based on the convex hull generated from the first possibility and the second possibility, and at least one of the user's medical interview information and attribute information. The apparatus according to claim 2.

9. The at least one processor is Based on the distribution information of accumulated data relating to at least one of the judgment image, the medical interview information, the attribute information, and the probability of disease, the reliability of the probability of disease is determined. The apparatus according to claim 1.

10. The at least one processor is A feature extractor for extracting predetermined features from one or more judgment images is input to a feature extractor for extracting predetermined features from one or more judgment images and the predetermined features are calculated. The distribution information of the accumulated data includes the feature quantities, and the reliability of the probability of the disease is determined. The apparatus according to claim 9.

11. The at least one processor is The one or more subject images are input to a trained judgment image selection model for selecting one or more judgment images from one or more subject images captured by the camera. The distribution information of the accumulated data includes the output results of the trained image selection model, and the confidence level of the possibility of the disease is determined. The apparatus according to claim 9.

12. The at least one processor is The distribution information of the accumulated data includes the output of the convolutional layer of the trained judgment model, and the confidence level of the possibility of the disease is determined. The apparatus according to claim 9.

13. The at least one processor is Based on the result of the assessment of the reliability of the likelihood of contracting the disease, it is decided whether or not to use the information regarding the likelihood of contracting the predetermined disease for further training of the trained judgment model. The apparatus according to claim 1.

14. The at least one processor is Information relating to the likelihood of contracting the aforementioned specified disease, which has been decided not to be used for further training of the aforementioned trained judgment model, is tagged. The apparatus according to claim 13.

15. The at least one processor is The information relating to the probability of contracting the aforementioned predetermined disease is stored in memory along with confidence information relating to the reliability of the probability of contracting the disease. The apparatus according to claim 1.

16. By being executed by at least one processor, A camera is used to capture images of the subject of the user, and one or more judgment images of the subject are acquired via the camera. The user's medical interview information and attribute information are obtained, Based on a trained assessment model for determining the likelihood of contracting a predetermined disease, and at least one of the assessment image, the medical history information, and the attribute information, the likelihood of contracting the predetermined disease is determined. Based on at least one of the aforementioned assessment image, the aforementioned medical interview information, and the aforementioned attribute information, the reliability of the likelihood of the disease is determined. A processing program that causes the aforementioned at least one processor to function in this manner.

17. A processing method that is performed by at least one processor, A step of acquiring one or more determination images of the subject via a camera used to capture images of the subject of the user, The steps include obtaining at least one of the user's medical interview information and attribute information, A trained judgment model for determining the likelihood of contracting a predetermined disease, and a step of determining the likelihood of contracting the predetermined disease based on at least one of the judgment image, the medical interview information, and the attribute information, A step of determining the reliability of the possibility of the disease based on at least one of the aforementioned judgment image, the aforementioned medical interview information, and the aforementioned attribute information, A processing method that includes this.

18. A photographic device equipped with a camera for taking images of the subject of the user, A processing device according to any one of claims 1 to 14, connected to the imaging device via a wired or wireless network, A processing system that includes this.

Images