Place of origin estimation system, place of origin estimation program, and place of origin estimation method

Abstract

This system provides a simple method for estimating a subject's place of origin by using readily available name information representing the subject's name as information to estimate the subject's place of origin. [Solution] The place of origin estimation system 100 includes a name information input means (input unit of the user terminal) that accepts name information of a target person from the user, a place of origin estimation means (AI server) that estimates the place of origin of the target person based on the name information, and a place of origin output means (display screen of the user terminal) that outputs the estimated place of origin. The place of origin estimation system also includes a user terminal, an AP server, and a generation AI server that constitute each of these means. The control unit in the user terminal, the control unit in the AP server, and the generation AI server 300 realize the place of origin estimation system by executing a place of origin estimation application, a place of origin estimation program, and a place of origin estimation model, respectively.

Description

Technical Field

[0001] The present invention relates to a place of origin estimation system, a place of origin estimation program, and a place of origin estimation method that can estimate the place of origin of a target person from the name of the target person.

Background Art

[0002] Conventionally, there have been various systems for estimating the nationality of a foreigner from voice information or face image information of the foreigner using a machine learning model. For example, Patent Document 1 below discloses a system for estimating the nationality of a foreigner related to voice information using a machine learning model that has learned the correlation between the voice information of a foreigner and the nationality. Also, Patent Document 2 below discloses a system for estimating the nationality of a foreigner related to a face image using a learned model that has learned the feature amounts of face images of people from various countries.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Patent Document 2

[0004] However, in each of the systems described in Patent Document 1 and Patent Document 2 above, voice data containing information sufficient to estimate at least the nationality of the target person or face image data containing sympathetic information is indispensable, and there is a problem that it is complicated to obtain the information necessary to estimate the nationality of the target person.

[0005] The present invention has been made to address the above problems, and an object thereof is to provide a place of origin estimation system that can easily estimate the place of origin of a target person by using name information representing the name of the target person, which is easily obtainable, as information for estimating the place of origin of the target person.

Summary of the Invention

[0006] To achieve the above objective, the present invention is characterized by a place of origin estimation system for estimating information about a subject's place of origin, comprising: a name information input means for inputting name information representing a subject's name, which includes at least the subject's name; a place of origin estimation means for estimating the subject's place of origin based on the name information input via the name information input means; and a place of origin output means for outputting the place of origin estimated by the place of origin estimation means.

[0007] According to this, the place of origin estimation system is configured to include a means for estimating the place of origin of a subject based on name information representing the subject's name, which includes at least the subject's given name. Therefore, the place of origin of a subject can be easily estimated based on readily available name information. Furthermore, experiments by the inventors have confirmed that the accuracy of estimating a subject's place of origin based only on information representing the subject's surname is about 50%, but the accuracy can be improved to 60-70% by estimating the subject's place of origin based on name information representing the subject's name, which includes at least the subject's given name.

[0008] Here, information regarding the subject's place of origin consists of at least the subject's place of origin. Here, the subject's place of origin means the place, region, country, or nationality where the subject was born, raised, or currently resides. In this case, the place of origin can be represented by a name such as a place name, region name, or country name, or by a mark, illustration, or national flag that represents these names. Furthermore, information regarding the subject's place of origin may also include information that represents the cultural, geographical, social, or historical characteristics or background of the subject's place of origin.

[0009] For example, information about a subject's place of origin may include information that represents the cultural, geographical, social, or historical characteristics or background of the subject's place of origin, such as the subject's native language, official language, or first language (hereinafter referred to as "understandable language"), the culture of the subject's place of origin (including academics, arts, manners, morals, customs, traditions, religion, and politics), or information useful when receiving the subject. Of this information about the subject's place of origin, information other than the subject's place of origin is called regional information.

[0010] Information useful when receiving a person (hereinafter referred to as "reception information") refers to information that would be helpful for the person receiving the person to know, such as precautions to take when receiving the person or information useful for building a good relationship with the person. For example, precautions to take when receiving the person include information that requires attention when receiving the person, such as manners, morals, customs, or religion of the person's place of origin. Information useful for building a good relationship with the person includes information about the values ​​or preferences of people in the person's place of origin, such as the preference for clearly expressing one's opinions or the fondness for alcohol.

[0011] Furthermore, the subject's name in this invention is a proper noun of the subject, consisting of at least the subject's name. This subject's name may also consist of the subject's surname and / or other names other than the subject's given name and surname (hereinafter referred to as "third name"). Specifically, the subject's name may consist of a combination of the subject's given name and surname, a combination of the subject's given name, third name and surname, a combination of the subject's surname and given name, a combination of the subject's third name and given name, a combination of the subject's third name, given name and surname, or a combination of the subject's given name and third name.

[0012] For example, names in countries such as Indonesia, Myanmar, Mongolia, and Afghanistan mainly consist of a given name only. Similarly, names in the Americas, Europe (excluding Hungary), Oceania, Central Asia, South Asia (excluding Iran), and Africa (excluding North Africa) mainly consist of a given name and surname, or a given name, third name, and surname. Furthermore, names in East Asia, Hungary, Singapore, and Vietnam mainly consist of a surname and given name. Finally, names in Southeast Asia (excluding Singapore and Vietnam) mainly consist of a given name only or a given name and surname.

[0013] Furthermore, personal names in North Africa mainly consist of a combination of a third name and a given name, or a combination of a given name and a surname. Similarly, personal names in Iran mainly consist of a combination of a given name and a surname, or a combination of a third name, a given name, and a surname. In West Asia and North Africa, personal names mainly consist of a combination of a given name, a third name, and a surname, or a combination of a given name and a third name. That is, the third name includes what is known as a middle name.

[0014] This third name consists of one or more names. For example, a third name may be a second name given to the subject in addition to their given name, a baptismal name, the names of the subject's ancestors such as their father, mother, or grandparents (including the given name and third name), a name derived from the father's or mother's name (including the given name and third name) (so-called patronymic or matrimony), the subject's nickname or common name, the subject's spouse's surname, the subject's maiden name, the name of a close friend of the subject or their relatives (including the given name and third name), the name of a person the subject respects (including the given name and third name), a title or honorific for the subject's status, the name of their birthplace, a tribal name, or the initials of these names.

[0015] Another feature of the present invention is that, in the above-mentioned place of origin estimation system, the name information is character information in which the name is written in characters, and the name is represented in a notation different from the formal spelling of the name.

[0016] According to this, the place of origin estimation system is configured to include a means for estimating the place of origin of a subject based on name information in which the subject's name is represented using a different character notation (hereinafter referred to as "unofficial notation") than the formal character notation (hereinafter referred to as "formal notation"). Therefore, even if the formal notation of the subject's name is unknown or the formal notation cannot be entered into the place of origin estimation system, the place of origin estimation system can estimate the subject's place of origin based on the subject's name represented by the unofficial notation. Note that "character notation" refers to the character notation of the subject's name.

[0017] Here, the official written spelling of a person's name refers to the spelling officially registered with the administrative body of the person's place of origin as the written representation of the person's name, or the spelling that the person most frequently uses in their daily life to represent their own name. For example, if person A, who has Japanese nationality, is pronounced "Yamada Taro," then "Yamada Taro," as registered in the family register for person A, is the official spelling. Similarly, if person B, who lives in the United States, is pronounced "Jennifer Van Stuart," and the spelling "Jennifer Van Stuart" is officially registered with the administrative body of the United States, then "Jennifer Van Stuart" is the official spelling for person B.

[0018] In this case, countries such as France, Germany, or Spain may use diacritical marks, or phonetic diacritical marks, to distinguish the pronunciation of identical characters. If the written form using these phonetic diacritical marks is registered with the government as the subject's name or is the form most frequently used by the subject, then the written form using the phonetic diacritical marks becomes the official form of the subject's name.

[0019] Furthermore, in the examples of subjects A and B, the subjects' names were shown using their full names, without any abbreviations. However, in some regions, nicknames that are partial omissions of full names or partial omissions of full names are also commonly used as names.

[0020] For example, some nicknames for Subject B include "Jennifer V. Stuart," "JV Stuart," or "Jennifer Stuart," which are abbreviations of part of her full name. Other nicknames for Subject B include "Jen / Jenn Van Stuart" or "Jenny / Jennie Van Stuart," which are derived from the first name "Jennifer" by using nicknames such as "Jen" or "Jenny."

[0021] In this case, if subject B most frequently uses "Jenny Stuart" as the written form of their name in daily life, then "Jenny Stuart" would also be considered an official spelling, in addition to "Jennifer Van Stuart," which is registered with U.S. government agencies.

[0022] On the other hand, a non-official spelling of a person's name refers specifically to a spelling, character, or written expression that differs from the official spelling of the person's name, either partially or entirely. For example, "Jennifer Van Stewart" and "Jennie Stuart," which use a different spelling for part of person B's name, "Jennifer Van Stewart" and "Jenny Stuart," which use different characters (katakana) for the entire name, or "Jennifer Van Stewart" and "Jenn V Stuart," which use different written expressions, are all considered non-official spellings.

[0023] In this case, "Jenn V Stuart" represents the name "Jennifer" of Subject B with the affectionate name "Jenn (Jen)", and represents the third name (so-called middle name) of Subject B with the initials omitting the comma, showing a different notation expression from the official notation of Subject B.

[0024] Also, when a pronunciation distinction symbol is used in the official notation, an alternative notation that represents the character with the pronunciation distinction symbol in a different notation expression corresponds to the unofficial notation. For example, as the unofficial notation of the character with the umlaut "¨" (hereinafter referred to as "e umlaut") attached to the English letter "e", the English letter "e" with the umlaut "¨" omitted or the notation that represents "e¨" as a character string in which the e umlaut is distinguished into an English letter and a pronunciation distinction symbol and the pronunciation distinction symbol is juxtaposed after the English letter corresponds to the unofficial notation.

[0025] In these official notations and unofficial notations, the case where the official notation of the name of the above-mentioned subject is unknown means, for example, that the user using the place of origin estimation system does not know the official notation of the name of the subject. Also, the case where the official notation of the name of the subject cannot be input means, for example, that the place of origin estimation system does not support the input operation of the characters used in the official notation or the user has not acquired the character notation of the official notation and cannot use it.

[0026] Also, another feature of the present invention is that in the place of origin estimation system, the name information is information representing the name in characters based on the name address recognized by the user using the place of origin estimation system.

[0027] According to this, the place of origin estimation system is configured to include a place of origin estimation means for estimating the place of origin of a target person from name information representing the name of the target person in characters based on the form of address of the name of the target person recognized by the user using the device. Therefore, the place of origin estimation system can estimate the place of origin of the target person from the name information based on the form of address of the name of the target person recognized by the user even when, for example, the user cannot accurately hear or read the name of the target person or when the voice input means of the place of origin estimation system cannot recognize the accurate form of address of the name of the target person.

[0028] Further, another feature of the present invention is that in the place of origin estimation system, the name information is composed of name information representing the name part of the target person in the name and non-name information representing the part other than the name of the target person in the name, and the place of origin estimation means estimates the place of origin based on the name information and the non-name information input via the name information input means.

[0029] According to this, the place of origin estimation system is configured to include a place of origin estimation means for estimating the place of origin of a target person based on name information representing the name part of the target person in the name of the target person and non-name information representing the part other than the name (hereinafter also referred to as "non-name part") in the entire name of the target person. Here, the non-name information is information representing the surname and / or the third name of the target person.

[0030] Therefore, the place of origin estimation system can improve the estimation accuracy by estimating the place of origin of the target person based on both the name information and the non-name information obtained by separating the name of the target person into the name part and the non-name part. According to the experiments of the present inventor, the estimation accuracy of the place of origin of the target person estimated based only on the name information before separation into the name information and the non-name information is 60 to 70%, but it has been confirmed that the estimation accuracy can be improved to 70 to 80% by estimating the place of origin of the target person based on both the name information and the non-name information after separating the name information.

[0031] Another feature of the present invention is that, in the above-mentioned place of origin estimation system, the place of origin estimation means estimates the place of origin using a machine learning model that has learned the relationship between the name of the subject and the place of origin of that subject.

[0032] According to this, the place of origin estimation system is configured to estimate a person's place of origin using a machine learning model that has learned the relationship between a person's name and their place of origin. Therefore, the place of origin estimation system can handle the estimation of places of origin based on a wider variety of names compared to a system that outputs (extracts) candidate places based on a person's name information from a database that associates the names of multiple people with candidate places that are candidates for each person's place of origin. Here, a wide variety of names include, for example, names not included in the aforementioned database, special names used within small groups such as limited ethnic groups, regions, or religions, or names that possess characteristics of multiple regions.

[0033] Another feature of the present invention is that, in the above-mentioned place of origin estimation system, the machine learning model is a large-scale language model.

[0034] According to this, the place of origin estimation system uses a large language model (LLM) to estimate a person's place of origin from their name. By taking into account the structure of the name and the meaning of the words used in it, the system can estimate the person's place of origin and improve the accuracy of the estimation.

[0035] Here, a large-scale language model is a language model that learns language patterns by performing deep learning on a vast amount of text data containing words and sentences spoken and written by humans, enabling it to understand the meaning and structure of sentences or generate sentences. This language model is a machine learning model that extracts patterns of natural language used by humans from text data and analyzes the probability of word occurrence to predict the word or phrase that will follow a particular word.

[0036] Another feature of the present invention is that the place of origin estimation system further includes a language estimation means that estimates a language that the subject can understand based on the place of origin estimated by the place of origin estimation means, and an estimated language output means that outputs the language estimated by the language estimation means.

[0037] According to this, the place of origin estimation system is configured to include a place of origin estimation means that estimates the place of origin of a subject based on name information, and a language estimation means that estimates the languages ​​the subject can understand based on the estimated place of origin. Therefore, the place of origin estimation system can easily estimate the place of origin and the languages ​​the subject can understand based on readily available name information.

[0038] Another feature of the present invention is that, in the place of origin estimation system, the language estimation means estimates multiple languages ​​as languages ​​that the subject can understand.

[0039] According to this, the place of origin estimation system is equipped with a language estimation means that estimates multiple languages ​​as understandable languages ​​for the subject, and therefore can present the user with multiple understandable languages ​​that the subject can comprehend.

[0040] Another feature of the present invention is that the place of origin estimation system further includes a reception information estimation means that estimates information useful when receiving a person based on the place of origin estimated by the place of origin estimation means, and a reception information output means that outputs the information estimated by the reception information estimation means.

[0041] According to this, the place of origin estimation system includes an interaction information estimation means that estimates useful information (interaction information) when interacting with a subject based on the place of origin estimated by the place of origin estimation means. Therefore, the place of origin estimation system can easily estimate the place of origin and interaction information of a subject based on readily available name information.

[0042] Furthermore, the present invention can be implemented not only as an invention of a place of origin estimation system, but also as an invention of a place of origin estimation program or a place of origin estimation method.

[0043] Specifically, the place of origin estimation program is a program that causes a computer device to perform a process to estimate information about the place of origin of a subject, and it is preferable that the computer device is caused to perform a name information input step to receive name information representing the subject's name, which consists of at least the subject's name; a place of origin estimation step to estimate the subject's place of origin based on the name information input via the name information input step; and a place of origin output step to output the place of origin estimated by the place of origin estimation step.

[0044] According to this, the place of origin estimation program can be expected to have similar effects and benefits to the invention of the place of origin estimation system described above.

[0045] Furthermore, the method for estimating a place of origin is a method for estimating a place of origin using a place of origin estimation system that estimates information about the place of origin of a subject, wherein the place of origin estimation system is composed of a place of origin estimation system described in any one of claims 1 to 9, and preferably includes a name information input step of inputting name information into a name information input means, a place of origin estimation step of causing a place of origin estimation means to estimate the place of origin of a subject based on the name information input into the name information input means, and a place of origin output step of causing the place of origin estimated by the place of origin estimation means to output to a place of origin output means.

[0046] According to this, the method for estimating place of origin can be expected to have the same effects as the invention of the place of origin estimation system described above. [Brief explanation of the drawing]

[0047] [Figure 1] This is a schematic block diagram showing the system configuration of the place of origin estimation system according to an embodiment of the present invention. [Figure 2]This is a flowchart showing the processing flow of the update program. [Figure 3] This is a flowchart showing the processing flow of the origin estimation app. [Figure 4] Figure 3 is a schematic diagram showing the initial screen displayed when the place of origin estimation app is run. [Figure 5] Figure 3 is a schematic diagram showing the menu screen that appears when the origin estimation app is run. [Figure 6] This flowchart shows the flow of the function implementation process. [Figure 7] Figure 6 is a schematic diagram showing the input screen that appears on the display screen after executing the function implementation process shown in Figure 6. [Figure 8] This is a flowchart showing the processing flow of the origin estimation program. [Figure 9] Figure 8 is a schematic diagram showing a portion of the estimation results screen displayed on the screen when the origin estimation program shown is executed. [Figure 10] Figure 8 is a schematic diagram showing another part of the estimation results screen displayed on the screen when the origin estimation program shown is executed. [Figure 11] This flowchart shows the processing flow of the survey processing program. [Modes for carrying out the invention]

[0048] (Configuration of the Origin Estimation System 100) Hereinafter, an embodiment of the place of origin estimation system according to the present invention will be described with reference to the drawings. Figure 1 is a block diagram illustrating the control system in the place of origin estimation system according to an embodiment of the present invention.

[0049] This place of origin estimation system 100 is an estimation device for estimating the place of origin of a subject based on the subject's name information. In this embodiment, the place of origin estimation system 100 is used when a user receives a subject in a medical institution such as a hospital or clinic. Here, name information is information representing the subject's name, consisting of at least the subject's name. Specifically, name information may include various types of information such as textual information representing the subject's name in formal or informal notation, audio information representing the pronunciation of the subject's name, or image information depicting the subject's name.

[0050] Here, "official spelling" refers to the spelling of the subject's name officially registered with the administrative body of the subject's place of origin, or the spelling that the subject most frequently uses in daily life to represent their own name. "Informal spelling" refers to a spelling, character, or written expression that differs from the official spelling, whether for part or all of the subject's name.

[0051] For example, if the official spelling of subject C is "Jennifer Van Stuart," then unofficial spellings include "Jennifer Van Stewart" (with a different spelling), "ジェニファ·バン·スチュアート" (using different characters / katakana), or "ジェニファ·ヴァン·スチュワート" and "Jenn Stuart" (with different written forms).

[0052] Here, "Jennifer Van Stewart" is one of several ways of writing Subject C's name in katakana, resulting from variations in spelling (for example, writing "Jennifer" as "Jennifa" and "Van" as "Ban"). Also, "Jenn Stuart" is a written representation of Subject C's name that differs from the official spelling, using the nickname "Jenn" for Subject C's given name "Jennifer" and omitting the third name (so-called middle name) "Van".

[0053] Of these informal spellings, "Jennifer Van Stewart" and "Jennifer Van Stewart" are written representations of Subject C's name based on its pronunciation. In other words, informal spellings of the subject's name include written representations of the subject's name based on its pronunciation.

[0054] Furthermore, the user is a person who uses the place of origin estimation system 100, and in this embodiment, this refers to a person who receives or assists in receiving the target person at a medical institution. Specifically, the user may be a doctor, nurse or technician, or other person who performs medical acts or acts equivalent to medical acts on users of a medical institution, or a person who performs medical administrative work. Furthermore, the target person is a person whose place of origin is estimated by the user using the place of origin estimation system 100. In this embodiment, the target person is a user of a medical institution, and specifically refers to a patient who visits a medical institution, a healthy person who undergoes a health checkup at a medical institution, or a visitor who visits a person hospitalized at a medical institution.

[0055] As shown in Figure 1, the origin estimation system 100 mainly comprises a user terminal 101, an application server 200 (hereinafter referred to as "AP server 200"), and a generation AI server 300. The user terminal 101 is a terminal device that receives name information of a target person from the user and displays the target person's estimated origin (hereinafter referred to as "estimated origin") based on the received name information, and is formed in a flat shape. In this embodiment, the user terminal 101 is a portable terminal device that the user can hold in their hand, specifically a known tablet terminal device. This user terminal 101 mainly comprises a storage unit 110, an input unit 120, a control unit 130, an output unit 140, and a communication unit 150.

[0056] The memory unit 110 is a storage device that stores the place of origin estimation application program (hereinafter referred to as the "place of origin estimation app"), the questionnaire processing program, and the OS (operating system) for operating the tablet terminal device. For example, the memory unit 110 is composed of an HDD, SSD, ROM, or RAM.

[0057] Furthermore, the storage unit 110 stores various information such as the name information of the target person received from the user by the user terminal 101 and the response information received from the AP server 200. In this case, the place of origin estimation application and the questionnaire processing program are stored in the storage unit 110 in advance by the user or administrator of the place of origin estimation system 100.

[0058] Here, the place of origin estimation app is an application program that works in conjunction with the AP server 200 to realize the place of origin estimation system 100. In this embodiment, the place of origin estimation app is composed of a web application (hereinafter referred to as "Web app") that operates on a web browser via the internet. This place of origin estimation app has a function to estimate the place of origin of a person based on the person's name information (hereinafter referred to as "place of origin estimation function") and a function to survey users' opinions on the estimation results estimated by the place of origin estimation function (hereinafter referred to as "survey function").

[0059] It should be noted that a place of origin estimation app does not necessarily have to be a web application. For example, a place of origin estimation app can be a native application that runs on the device (hereinafter referred to as "native app") or a hybrid application that runs on the device and is developed using a web application development language.

[0060] Furthermore, the questionnaire processing program is a computer program for conducting a questionnaire to investigate the user's opinion on the estimation results estimated by the place of origin estimation system 100. Specifically, the questionnaire processing program controls the operation of the control unit 130 to output the questionnaire to the output unit 140 and receives the user's response to the questionnaire via the input unit 120. The questionnaire processing program also controls the operation of the control unit 130 to generate feedback information using the responses received from the user, and then outputs the user's response and the feedback information as questionnaire result information to the AP server 200 (storage unit 210) via the communication unit 150.

[0061] In this embodiment, the questionnaire results information includes birthplace accuracy information that evaluates the accuracy of the estimated birthplace, region-related accuracy information that evaluates the accuracy of the estimated region-related information, comment information which is the subject's comment on the estimation results, actual birthplace information which represents the subject's true birthplace, actual region-related information which represents the subject's true region-related information, and feedback information. Here, feedback information is information used to reflect the user's evaluation of the estimation results in the birthplace estimation system 100. Specifically, the feedback information includes name-birthplace comparison information which associates the subject's name information with actual birthplace information, and actual birthplace-related information which associates actual birthplace information with actual region-related information.

[0062] Furthermore, the response information is information about the response output by the AP server 200 in response to the instruction of the control unit 130. In this embodiment, the response information includes initial information output by the AP server 200 in response to the notification to start the place of origin estimation application, estimated place of origin information (hereinafter also referred to as "estimated place of origin information") output by the AP server 200 to the control unit 130 in response to the instruction to execute the place of origin estimation function, and estimated region-related information (hereinafter referred to as "estimated region-related information").

[0063] This initial information includes, for example, initial screen information related to the initial screen of the place of origin estimation app, menu screen information related to the menu screen where the operation items are displayed in a list, input screen information related to the input screen where the user enters the name information of the target person, estimation result screen information related to the estimation result screen for displaying the estimation results, and survey screen information related to the survey screen where a questionnaire presented to the user is displayed. Here, the initial screen is the operation screen that is displayed first when the place of origin estimation app is run.

[0064] The input unit 120 is an input device for receiving user operations on the place of origin estimation system 100, and outputs operation information corresponding to the user's input to the control unit 130. In this embodiment, the input unit 120 is implemented by being displayed on a touch panel that constitutes a tablet terminal device. The input unit 120 outputs the name information of the subject received from the user to the storage unit 110 according to the instructions of the control unit 130. In this case, the input unit 120 receives the name information of the subject as two independent pieces of information: name information representing the part of the subject's name and non-name information representing the part of the subject's name other than their name (specifically, their surname and / or third name). This input unit 120 corresponds to the name information input means according to the present invention.

[0065] The control unit 130 is composed of a microcomputer consisting of a CPU, ROM, RAM, etc., and controls the operation of each part of the user terminal 101 by executing the place of origin estimation application stored in the memory unit 110. Specifically, the control unit 130, in response to the input of the subject's name information from the input unit 120, instructs the AP server 200 to estimate the subject's place of origin based on the subject's name information. The control unit 130 also displays the response information transmitted by the AP server 200 in response to the instruction on the output unit 140.

[0066] Furthermore, the control unit 130 executes the survey processing program stored in the memory unit 110 to display the survey screen, acquire survey results, generate feedback information, and generate survey result information, respectively. Specifically, the control unit 130 includes a survey processing unit 131 that operates when the survey processing program is executed.

[0067] The survey processing unit 131 accepts the user's responses to the survey via the input unit 120 by displaying the survey screen information stored in the storage unit 110 on the output unit 140, in accordance with instructions from the user. The survey processing unit 131 also generates feedback information using the subject's name information stored in the storage unit 110 and the user's responses, and then outputs the user's responses and feedback information as survey result information to the AP server 200 (storage unit 210) via the communication unit 150. Specifically, the survey processing unit 131 generates name-place-of-origin matching information by associating the subject's name information with the place-of-origin information included in the user's responses, and place-of-origin related information by associating the place-of-origin information with the region-related information included in the user's responses.

[0068] The output unit 140 is an output device for outputting response information stored in the storage unit 110 in accordance with instructions from the control unit 130. This output unit 140 is mainly configured to include a display screen 141.

[0069] The display screen 141 is a display device for displaying response information as an image based on instructions from the control unit 130. In this embodiment, the display screen 141 is composed of a touch panel display that constitutes a tablet terminal device. This output unit 140 corresponds to the place of origin output means, estimated language output means, and reception information output means according to the present invention.

[0070] The communication unit 150 is an electrical circuit that sends and receives information to and from the AP server 200 via a network NW, and its operation is controlled by the control unit 130. In this embodiment, the communication unit 150 outputs the name information of the subject received from the user by the input unit 120 and the survey result information generated by the survey processing unit 131 to the AP server 200 (storage unit 210), respectively, according to instructions from the control unit 130. The communication unit 150 also outputs the response information received from the AP server 200 to the storage unit 110 according to instructions from the control unit 130.

[0071] Furthermore, the user terminal 101 includes a power supply unit (not shown) that supplies power to the storage unit 110, input unit 120, control unit 130, output unit 140, and communication unit 150, respectively. In this embodiment, the power supply unit is composed of a battery (a rechargeable secondary battery such as a nickel-metal hydride battery or a lithium-ion battery) that stores power from an external power source such as a household power supply (100V). However, since the power supply unit only needs to supply power to the user terminal 101, it may also be configured to receive power directly from a primary battery such as a dry cell battery or from a household power supply (100V) and supply power to each unit.

[0072] The AP server 200 is a computer device that, in accordance with instructions from the control unit 130, works in cooperation with the generation AI server 300 to estimate the place of origin of a person based on the name information of the person received from the user. The AP server 200 mainly consists of a storage unit 210, a control unit 230, and a communication unit 250.

[0073] The memory unit 210 is a storage device that stores the place of origin estimation program, the update processing program, and the OS (operating system) for running the AP server 200. For example, the memory unit 210 is composed of an HDD, SSD, ROM, or RAM. The memory unit 210 also stores initial information to be output to the memory unit 110 in response to a notification to start the place of origin estimation application, various information for executing the questionnaire processing program, questionnaire result information, various information for executing the place of origin estimation program, and various information for executing the update processing program. In this case, the place of origin estimation program and the update processing program are stored in the memory unit 210 in advance by the user or administrator of the place of origin estimation system 100.

[0074] Here, the various pieces of information necessary to execute the survey processing program include, for example, survey information, survey screen information, place of origin information, and region-related information. This survey information is information for generating survey screen information that represents the survey screen displayed on the display screen 141 of the user terminal 101. Here, the survey screen is an image displaying a survey to investigate the user's opinion on the estimation results estimated by the place of origin estimation function.

[0075] In this embodiment, the questionnaire screen is configured to include a question item asking whether the estimated place of origin is correct or incorrect, a question item asking whether the estimated languages ​​understood by the subject are correct or incorrect, a question item asking whether the estimated contact information is correct or incorrect, a question item asking for the name of the subject corresponding to the incorrect estimated place of origin and that subject's true place of origin, a question item asking for the name of the subject corresponding to the incorrect estimated region-related information and that subject's true region-related information, and a comment field for freely receiving the user's opinion.

[0076] Furthermore, the place of origin estimation program is a computer program that controls the operation of the control unit 230 to estimate the place of origin of a subject based on their name using the generating AI server 300. Specifically, the place of origin estimation program controls the operation of the control unit 230 to generate an input statement instructing the generating AI server 300 to estimate the place of origin of the subject. The place of origin estimation program also controls the operation of the control unit 230 to generate estimation result presentation information using the response to the input statement received from the generating AI server 300 to present the estimated place of origin and estimated region-related information to the user as estimation results.

[0077] Here, the input sentence is the character information input to the machine learning model provided by the generation AI server 300, and is a so-called prompt. This input sentence mainly consists of a main sentence and an appendix. The main sentence is the part of the input sentence that indicates instructions to the generation AI server 300, or, in addition to the instructions, constraints on the response that the generation AI server 300 will output. The appendix is ​​the part of the input sentence that indicates the name information of the subject. In this case, the input sentence is generated by the control unit 230 and then output to the generation AI server 300 according to the instructions of the control unit 230.

[0078] Furthermore, the various information required to execute the place of origin estimation program includes, for example, the subject's place of origin, regional information, the subject's name information received from the control unit 130, input sentence generation information for generating input sentences, the subject's full name information generated using the subject's given name and non-given name information, input sentence information representing the input sentence generated based on the input sentence generation information, response information representing the response output by the generation AI server 300 in response to the input sentence, place of origin related information associating the subject's place of origin with the regional information, and estimation result presentation information generated using the response information and the place of origin related information. In this case, the subject's place of origin, regional information, input sentence generation information, and place of origin related information are stored in the storage unit 210 in advance by the user or administrator of the place of origin estimation system 100.

[0079] In this embodiment, the subject's place of origin is represented by the name of the place of origin and the national flag representing that name. In addition, in this embodiment, the regional information consists of understandable language information representing the languages ​​the subject understands, reception information, and religious information regarding religion.

[0080] The input sentence generation information is character information used as a template for the input sentence when the control unit 230 generates the input sentence, and mainly consists of main sentence generation information and supplementary sentence generation information. This main sentence generation information is character information used as a template when generating the main sentence in the input sentence, and in this embodiment, it is configured to include constraints that limit the output format of the response generated by the generation AI server 300. Specifically, the main sentence generation information is configured to include a condition that limits the response to be selected from multiple candidates specifying the subject's place of origin. The supplementary sentence generation information is character information used as a template when generating supplementary sentences in the input sentence.

[0081] Furthermore, the input sentence information mainly consists of main sentence information. This main sentence information is generated by the control unit 230 by adding the additional sentence information, which is generated using the subject's name information and additional sentence generation information, to the main sentence generation information.

[0082] Furthermore, the update processing program is a computer program that controls the operation of the control unit 230 to perform fine tuning of the machine learning model (hereinafter referred to as the "place of origin estimation model") which performs the process of estimating the place of origin of a subject from the subject's name information. Here, fine tuning refers to the technique of adjusting a trained model by retraining part or all of the trained model, which has been trained by pre-training, using new additional data. The update processing program is also a computer program that controls the operation of the control unit 230 to update information on additional data (hereinafter referred to as "additional data information") and place of origin-related information for retraining the place of origin estimation model using the questionnaire result information generated by the questionnaire processing unit 131.

[0083] Furthermore, the various information required to execute the update processing program includes, for example, information on additional data for retraining the place of origin estimation model (hereinafter referred to as "additional data information") and update information regarding whether or not the additional data information has been updated. In this embodiment, the additional data information consists of correct / incorrect information (place of origin correct / incorrect information and region-related correct / incorrect information) and feedback information (name-place of origin comparison information) included in the survey results information.

[0084] The various information necessary to execute these update processing programs may be directly input into the storage unit 210 by the user or administrator using an input device (not shown) on the AP server 200 or an input device connected to the AP server 200 via a network NW, or it may be configured so that the information is automatically input into the storage unit 210 from an external device via the network NW when the update processing program is executed.

[0085] Specifically, the update processing program can be configured to periodically collect information from the network NW by crawling and input it into the storage unit 210 as a process for storing various information necessary for executing the update processing program in the storage unit 210. Here, crawling refers to the process in which software automatically patrols the internet and collects and saves the content of web pages (i.e., web content) from various websites.

[0086] The control unit 230 is composed of a microcomputer consisting of a CPU, ROM, RAM, etc., and executes the origin estimation program stored in the memory unit 210 to generate the input sentence and the estimation result presentation information, respectively. Specifically, the control unit 230 includes an input sentence generation unit 231, an answer acquisition unit 232, and a presentation information generation unit 233, which are operated by the execution of the origin estimation program.

[0087] The input sentence generation unit 231 generates an input sentence containing instructions for the place of origin estimation model, using the subject's name information and input sentence generation information pre-stored in the storage unit 210, in accordance with instructions from the control unit 130. In this embodiment, the input sentence generation unit 231 generates an input sentence by adding the subject's name information and non-name information, respectively, to the input sentence generation information, which is a template for the input sentence. Specifically, the input sentence generation unit 231 generates additional sentence information by inserting part or all of the subject's name information into the additional sentence generation information, and then completes the input sentence by inserting the generated additional sentence information into the main sentence generation information.

[0088] The response acquisition unit 232 stores the received response information in the storage unit 210 in response to receiving response information from the generation AI server 300 via the communication unit 250.

[0089] The presentation information generation unit 233 generates estimated result presentation information using the response information and place of origin related information stored in the storage unit 210. In this embodiment, the presentation information generation unit 233 extracts (searches) estimated region related information (specifically, estimated understandable language information and reception information) corresponding to the estimated place of origin from the place of origin related information to generate estimated result presentation information. In this case, the presentation information generation unit 233 extracts multiple languages ​​as languages ​​that the subject can understand from the understandable language information that constitutes the place of origin related information to generate estimated result presentation information.

[0090] Furthermore, the presentation information generation unit 233 generates estimation result screen information for displaying the estimation result on the display screen 141 using the input screen information, response information, and generated estimation result presentation information stored in the storage unit 210. This presentation information generation unit 233 corresponds to the language estimation means and the response information estimation means according to the present invention.

[0091] In addition, the control unit 230 performs update processing of additional data information and place of origin-related information, as well as fine-tuning of the place of origin estimation model, by executing an update processing program stored in the memory unit 210. Specifically, the control unit 230 includes an update processing unit 234 that is activated by the execution of the update processing program.

[0092] The update processing unit 234 updates the additional data information using at least one of the following: place of origin accuracy information, region-related accuracy information, and feedback information (specifically, name-place of origin comparison information) included in the survey results information. After updating, it generates update information indicating that the additional data information has been updated and stores it in the storage unit 210. The update processing unit 234 also updates the place of origin-related information using the feedback information (specifically, this place of origin-related information) included in the survey results information and stores it in the storage unit 210. In addition, in response to the update of the additional data information, the update processing unit 234 inputs the latest additional data information into the place of origin estimation model and instructs it to perform fine tuning of the place of origin estimation model.

[0093] The communication unit 250 is an electrical circuit that transmits and receives information between the user terminal 101 and the generation AI server 300 via a network NW, and its operation is controlled by the control unit 230. In this embodiment, the communication unit 250 transmits the input sentence information generated by the input sentence generation unit 231 to the generation AI server 300 in accordance with instructions from the control unit 230, and receives the response information output by the generation AI server 300 as a response to the input sentence information. The communication unit 250 also transmits additional data information stored in the storage unit 210 to the generation AI server 300 in accordance with instructions from the control unit 230.

[0094] The generation AI server 300 is a computer device that, in response to a request from the user terminal 101, controls the operation of a place of origin estimation model that has learned the relationship between a person's name and the place of origin of the person with that name, and estimates the place of origin of the person based on the name information of the person received from the AP server 200. This generation AI server 300 is mainly configured to include a place of origin estimation model. This generation AI server 300 corresponds to the place of origin estimation means according to the present invention.

[0095] A place of origin estimation model is a machine learning model that uses information showing the relationship between a person's name and the place of origin of that person as training data, so that it can estimate the place of origin of a specific person based on the input name. In this case, the place of origin estimation model is a machine learning model for natural language processing (NLP), and specifically, it is composed of a large-scale language model (LLM). Here, natural language processing refers to the technique of having a computer device process natural language used by humans according to a purpose.

[0096] In this embodiment, the origin estimation model uses ChatGPT® developed by OpenAI. However, the origin estimation model is not necessarily limited to ChatGPT® and other large-scale language models can be used. For example, the origin estimation model can consist of Google® Gemini, Microsoft® Copilot, Anthropic® Claude, or Perplexity AI's Perplexity.

[0097] (Activation of the place of origin estimation system 100) Next, the operation of the origin estimation system 100 configured in this way will be explained. In this embodiment, the origin estimation system 100 is used when a person (user) who handles visitors at a medical institution greets a person (target) who has come to the medical institution. In this case, the user or administrator fine-tunes the origin estimation model provided by the generation AI server 300 in advance by executing an update processing program on the AP server 200.

[0098] Specifically, the user or administrator first instructs the control unit 230 (update processing unit 234) in the AP server 200 to execute the update processing program. In response to this instruction, the control unit 230 (update processing unit 234) starts executing the update processing program shown in Figure 2 in step S100, and in step S102 determines whether or not new survey result information has been acquired (stored).

[0099] In this case, if the control unit 230 (update processing unit 234) has survey result information stored in the storage unit 210, it determines "Yes" as to whether new survey result information has been acquired and proceeds to step S104. On the other hand, if the control unit 230 (update processing unit 234) has not had survey result information stored in the storage unit 210, it determines "No" in this determination process and proceeds to step S106.

[0100] Next, in step S104, the control unit 230 (update processing unit 234) updates the additional data information and place of origin-related information stored in the storage unit 210 using the survey result information, and then deletes the survey result information from the storage unit 210.

[0101] In this embodiment, the control unit 230 (update processing unit 234) updates the additional data information using the correct / incorrect information (place of origin correct / incorrect information and region-related correct / incorrect information) and feedback information (specifically, name-place of origin comparison information) included in the survey results information. In this case, the control unit 230 (update processing unit 234) generates update information indicating that the additional data information has been updated and stores it in the storage unit 210 upon completion of the update of the additional data information. The control unit 230 (update processing unit 234) also updates the place of origin-related information using the feedback information (specifically, place of origin-related information) included in the survey results information.

[0102] Next, in step S106, the control unit 230 (update processing unit 234) determines whether or not the additional data information has been updated. In this case, if the update information generated in step S104 indicates "update available", the control unit 230 (update processing unit 234) determines that the additional data information has been updated and determines "Yes", then proceeds to step S108. In step S108, the control unit 230 (update processing unit 234) instructs the generation AI server 300 to perform fine tuning of the place of origin estimation model using the updated additional data information.

[0103] On the other hand, if the update information indicates "no update", the control unit 230 (update processing unit 234) determines "No" in the determination process in step S106 and proceeds to step S110, ending the execution of this update processing program. In other words, if the additional data information has not been updated, the control unit 230 (update processing unit 234) does not perform the fine-tuning process of the place of origin estimation model.

[0104] Next, the user instructs the control unit 130 to launch the place of origin estimation application. Specifically, the user displays a shortcut icon for the place of origin estimation application on the display screen 141 of the user terminal 101, and then taps the icon. In response to this instruction, the control unit 130 starts executing the place of origin estimation application shown in Figure 3 in step S200. Next, in step S202, the control unit 130 stores the response information (specifically, initial information) output by the control unit 230 in response to the notification to launch the place of origin estimation application in the storage unit 110.

[0105] Next, in step S204, the control unit 130 displays the initial screen W141 of the place of origin estimation application on the display screen 141 based on the initial screen information included in the initial information received from the control unit 230, as shown in Figure 4. In this embodiment, the initial screen W141 displays a menu button 141a in the upper left portion of the display screen 141 for transitioning the displayed content to the menu screen W142.

[0106] Next, in step S206, the control unit 130 executes a termination subprogram. This termination subprogram is a computer program that forcibly terminates the running place of origin estimation application according to the user's operation. Specifically, the control unit 130 terminates the execution of the place of origin estimation application by swiping the content displayed on the display screen 141 (the displayed image resulting from the execution of the place of origin estimation application) off the display screen 141. In this case, by executing the termination subprogram, the control unit 130 can forcibly terminate the execution of the place of origin estimation application according to the user's operation, even if other processing is being carried out in the place of origin estimation application. This termination subprogram is always executed while the place of origin estimation application is running.

[0107] Next, in step S208, the control unit 130 determines whether the menu button 141a displayed on the display screen 141 has been selected. In this case, the control unit 130 continues to determine "No" in this determination process until the user taps the menu button 141a, waiting for the user to tap the menu button 141a. If the user taps the menu button 141a, the control unit 130 determines "Yes" in this determination process and proceeds to step S210. In step S210, the control unit 130 controls the display screen 141 to display the menu screen W142.

[0108] In this embodiment, as shown in Figure 5, the control unit 130 displays the menu screen W142 on top of the initial screen W141 already displayed on the display screen 141, based on the menu screen information stored in the storage unit 110. Specifically, the control unit 130 displays the menu screen W142 on top of the left area when the initial screen W141 is divided into left and right sections. In this case, the menu screen W142 displays the origin estimation button 142a and the survey button 142b, respectively, which are used to instruct the user to activate the origin estimation function and the survey function provided by the origin estimation application.

[0109] Next, in step S212, the control unit 130 executes the menu screen back subprogram. This menu screen back subprogram is a computer program that returns the display content of the display screen 141 from the menu screen W142 (the currently displayed screen) to the initial screen W141 (the previous screen) according to the user's operation.

[0110] Specifically, the control unit 130 starts executing the menu screen back subprogram in step S212 and determines whether an operation to return to the initial screen W141 has been performed on the input unit 120. In this case, if a part of the initial screen W141 displayed to the right of the menu screen W142 on the display screen 141 (the dot hatching part in Figure 5) is tapped, the control unit 130 determines that an operation to return to the initial screen W141 has been performed and returns to step S208. On the other hand, if the part of the initial screen W141 is not tapped, the control unit 130 determines "No" in this determination process and proceeds to step S214. This menu screen back subprogram is always executed while the place of origin estimation application is running.

[0111] Next, in step S214, the control unit 130 executes function implementation processing to realize the place of origin estimation function and the questionnaire function, respectively. Specifically, the control unit 130 first starts executing the function implementation processing shown in Figure 6 in step S300, and in step S302, it determines whether the place of origin estimation button 142a displayed on the menu screen W142 has been selected. In this case, if the place of origin estimation button 142a displayed on the menu screen W142 has been tapped by the user, the control unit 130 determines that the place of origin estimation button 142a has been selected and determines "Yes", and proceeds to step S304.

[0112] If the origin estimation button 142a is selected in step S302, the control unit 130 displays the input screen W143 on the display screen 141 in step S304 based on the input screen information stored in the storage unit 110, as shown in Figure 7. In this embodiment, the input screen W143 consists of a name input field 143A for entering the name of the subject and an estimation result display field 143B for displaying the estimation result.

[0113] In this case, the input screen W143 displays the message "Please enter the name of the person whose nationality you want to know" at the top of the screen, explaining how to use the origin estimation function to the user. Below this message are the name input field 143A and the estimation result display field 143B. Additionally, the input screen W143 has a back button 143e at the top left to return to the menu screen W142.

[0114] This name input field 143A consists of a given name input field 143a and a non-name input field 143b. The given name input field 143a is for entering the given name portion of the subject's entire name. The non-name input field 143b is for entering the parts of the subject's entire name other than their given name (specifically, their surname and / or third name).

[0115] Furthermore, the estimated result display area 143B consists of an estimated place of origin area 143c and an estimated region-related information area 143d. The estimated place of origin area 143c is a field for displaying the estimated place of origin, which constitutes the estimated result presentation information. In this case, on the input screen W143 where the subject's name has not yet been entered, the message "Please enter the name." prompting the user to enter the subject's name is displayed in the estimated place of origin area 143c. The estimated region-related information area 143d is a field for displaying estimated region-related information, which constitutes the estimated result presentation information. In this case, the estimated region-related information area 143d consists of a language area 143da for displaying the subject's understandable languages ​​and a reception information area 143db for displaying reception information.

[0116] Next, in step S306, the control unit 130 executes the input screen back subprogram. This input screen back subprogram is a computer program that returns the content displayed on the display screen 141 from the unentered input screen W143 (the currently displayed screen), the input screen W143 after input, or the estimated result screen W144 to the menu screen W142 (the previous screen) according to the user's operation.

[0117] Specifically, the control unit 130 starts executing the input screen back subprogram in step S306 to determine whether an operation to return to the menu screen W142 has been performed on the input unit 120. In this case, if the back button 143e displayed on the input screen W143 is tapped by the user, the control unit 130 determines "Yes" as an operation to return to the menu screen W142 has been performed and returns to step S302. On the other hand, if the back button 143e is not tapped, the control unit 130 determines "No" in this determination process and proceeds to step S308. This input screen back subprogram is always executed while the function implementation process is running.

[0118] Next, in step S308, the control unit 130 determines whether or not character information has been entered into the name input field 143A. In this case, the control unit 130 continues to determine "No" in this determination process until the user starts an input operation in the name input field 143A, and waits for the user to start an input operation in the name input field 143A. Then, when the user starts an input operation in the name input field 143A, the control unit 130 determines "Yes" in this determination process and proceeds to step S310. Next, in step S310, the control unit 130 displays character information corresponding to the user's input operation in the name input field 143A of the input screen W143. In this embodiment, the control unit 130 displays character information corresponding to the user's input operation in the name input field 143a and the non-name input field 143b that constitute the name input field 143A.

[0119] Next, in step S312, the control unit 130 determines whether the input completion button displayed on the display screen 141 has been selected by the user. Here, the input completion button is a button that notifies the control unit 130 that the user has completed the input operation of the subject's name, and is also a button that instructs the control unit 130 to estimate the subject's place of origin. In this embodiment, the input completion button is the Enter key of the keyboard displayed on the display screen 141 when the subject's name is entered.

[0120] In this case, the control unit 130 continues to determine "No" in this determination process until the user taps the Enter key displayed on the display screen 141, and returns to step S308. As a result, the control unit 130 repeatedly executes each of the processes from steps S308 to S312 until the user has finished entering the name of the target person. On the other hand, if the user taps the Enter key, the control unit 130 determines "Yes" in this determination process and proceeds to step S314. The name information input process from steps S308 to S312, from receiving the input of the target person's name information to completion, corresponds to the name information input step according to the present invention.

[0121] Next, in step S314, the control unit 130 stores the name of the subject received from the user as subject name information in the storage unit 210 via the communication unit 150. In this embodiment, we will explain assuming that the user enters "Mariel" in the name input field 143a of the input screen W143 as the name of subject D, "Mariel Raymundo Cruz" (full name and formal spelling), and also enters "Raymond Cruz" in the non-name input field 143b. In this case, the subject name information "Mariel" and non-name information "Raymond Cruz" entered by the user are information that represents the name of subject D in katakana and symbols (characters) based on the pronunciation of subject D's name as recognized by the user.

[0122] In this case, the control unit 130 stores "Mariel" as the name information of subject D in the memory unit 210, and also stores "Raymond Cruz" separately in the memory unit 210 as the non-name information of subject D. The control unit 130 also sends an instruction (hereinafter referred to as the "place of origin estimation instruction") to the control unit 230 requesting that the place of origin of the subject be estimated based on the name information and non-name information stored in the memory unit 210.

[0123] In response to this instruction to estimate the place of origin, the control unit 230 starts executing the place of origin estimation program shown in Figure 8 in step S400. Next, in step S402, the control unit 230 (input sentence generation unit 231) generates an input sentence to be input to the generation AI server 300 based on the place of origin estimation instruction received from the control unit 130. In this embodiment, the control unit 230 (input sentence generation unit 231) generates an input sentence using the name information "Mariel" and non-name information "Raymond Cruz" of the subject D stored in the storage unit 210, and the input sentence generation information stored in the storage unit 210.

[0124] First, the control unit 230 (input sentence generation unit 231) generates additional sentences that constitute the input sentence. Specifically, the control unit 230 (input sentence generation unit 231) retrieves the name information of subject D, "Mariel", the non-name information of subject D, "Raymond Cruz", and the additional sentence generation information, "In the name of subject D, the first name is..." from the storage unit 210. <x1>That's it. The parts other than the given name (for example, surname or middle name) are... <x2>It reads out "It is." Next, the control unit 230 (input sentence generation unit 231) inserts the name information "Mariel" and non-name information "Raymond Cruz" of the subject D into the "X1" and "X2" parts of the added sentence generation information, respectively.

[0125] Through these processes, the control unit 230 (input sentence generation unit 231) can generate the additional sentence information "The subject's first name is <Mariel>. The other part (for example, surname or middle name) is <Raymond Cruz>."

[0126] Next, the control unit 230 (input sentence generation unit 231) receives the main sentence generation information from the storage unit 210: "Please estimate and list the nationality that is closest to the nationality of the subject with the following name from the following nationalities. Please answer with only the country name. Also, the reason for the estimation is not required in the answer." <y1>> The nationality is read out as "Japan, Europe, the United States, China, Brazil, the Philippines, Indonesia, Vietnam, Other". Next, the control unit 230 (input sentence generation unit 231) inserts the generated additional sentence information into the "Y1" portion of the main sentence generation information to generate the main sentence information. Then, the control unit 230 (input sentence generation unit 231) stores the generated main sentence information as input sentence information in the storage unit 210.

[0127] Through these processes, the control unit 230 (input sentence generation unit 231) can generate the input sentence information: "Please estimate and list the nationality that is closest to the nationality of the subject with the following name from the following nationalities. Please answer with only the country name. You do not need to give a reason for your estimation. <<The subject's first name is <Mariel>. The other part (e.g., surname or middle name) is <Raymond Cruz>> Nationality: Japan, Europe / America, China, Brazil, Philippines, Indonesia, Vietnam, Other."

[0128] Next, in step S404, the control unit 230 transmits the generated input sentence information to the generation AI server 300 via the communication unit 250. That is, the control unit 230 transmits input sentence information including name information and non-name information, in which the subject's name has been separated into a name part and a non-name part, to the generation AI server 300.

[0129] Next, the generating AI server 300 controls the place of origin estimation model according to the input sentence information received from the control unit 230 to estimate the place of origin of the person corresponding to the input sentence information and generate a response. In this embodiment, the place of origin estimation model estimates the place of origin of the person based on the given name information and non-given name information of the person indicated in the input sentence information, in accordance with constraints on the response included in the input sentence information. This process of estimating the place of origin of the person based on the given name information and non-given name information corresponds to the place of origin estimation step according to the present invention.

[0130] In this embodiment, the place of origin estimation model selects the subject's nationality from among a plurality of countries specified in the input sentence information (Japan, Europe, the United States, China, Brazil, the Philippines, Indonesia, Vietnam, or others) and generates a response in the form of a country name. Specifically, the place of origin estimation model estimates "Philippines" as the place of origin (nationality) of subject D based on the given name information "Mariel" and non-given name information "Raymond Cruz" and generates a response. Next, the generating AI server 300 transmits the response "Philippines" to the input sentence information as response information to the control unit 230.

[0131] Next, in step S406, the control unit 230 (answer acquisition unit 232) determines whether or not it has acquired the answer information. In this case, the control unit 230 (answer acquisition unit 232) continues to determine "No" in this determination process until the communication unit 250 receives the answer information from the generation AI server 300, and waits until the communication unit 250 receives the answer information. Then, if the communication unit 250 has received the answer information, the control unit 230 (answer acquisition unit 232) determines "Yes" in this determination process and stores the answer information in the storage unit 210 in step S408.

[0132] Next, in step S410, the control unit 230 (presentation information generation unit 233) generates estimated result presentation information using the answer information and place of origin related information stored in the storage unit 210. In this case, the control unit 230 (presentation information generation unit 233) extracts estimated region related information corresponding to the answer information from the place of origin related information to generate estimated result presentation information. In this embodiment, the control unit 230 (presentation information generation unit 233) extracts understandable language and reception information (i.e., estimated region related information) corresponding to the Philippines (estimated place of origin) from the place of origin related information to generate estimated region related information.

[0133] Next, in step S412, the control unit 230 (presentation information generation unit 233) generates estimation result screen information using the input screen information stored in the storage unit 210, the response information generated by the place of origin estimation model, and the estimation result presentation information. In this embodiment, the control unit 230 (presentation information generation unit 233) generates estimation result screen information by inserting the estimation result presentation information into the estimation result display field 143B of the input screen information.

[0134] Next, in step S414, the control unit 230 transmits the generated estimation result screen information to the control unit 130 via the communication unit 250. After that, in step S416, the control unit 230 terminates the execution of the place of origin estimation program and proceeds to step S316 in the function implementation process.

[0135] Next, in step S316, the control unit 130 determines whether or not it has acquired the estimated result screen information. In this case, the control unit 130 continues to determine "No" in this determination process until the communication unit 150 receives the estimated result screen information from the control unit 230, and returns to step S308. As a result, the control unit 130 repeatedly executes each of the processes from steps S308 to S316 until it acquires the estimated result screen information from the control unit 230 via the communication unit 150. On the other hand, if the communication unit 150 has received the estimated result screen information from the control unit 230, the control unit 130 determines "Yes" in this determination process and proceeds to step S318.

[0136] Next, in step S318, the control unit 130 displays the estimation result screen W144 on the display screen 141 based on the estimation result screen information received from the control unit 230, as shown in Figures 9 and 10, respectively. In this embodiment, the estimation result screen W144 is composed of a name input field 143A and an estimation result display field 143B, similar to the input screen W143.

[0137] In this case, the name input field 143A displays the name information that was displayed in the name input field 143a and the non-name input field 143b respectively in response to the user's input operation in step S310. In addition, the estimated result display field 143B displays "Philippines" as the name of the estimated nationality in the estimated place of origin field 143c, and the Philippine flag, "Tagalog, English" and "Tagalog and English may be spoken" are displayed in the language field 143da.

[0138] Furthermore, the reception information section 143db displays the following points to note when serving person D: "1. Direct pointing: Directly pointing at a person is considered rude in Filipino culture. 2. Waving with palm down: The gesture of waving with palm down to call someone over can be interpreted in the Philippines as a gesture used to call animals, and is considered inappropriate when used towards people. 3. Calling out in a loud voice: Calling out to someone in a loud voice is considered rude in the Philippines, as in many other cultures." The process of displaying the estimated results obtained by executing this place of origin estimation program on the display screen 141 of the user terminal 101 corresponds to the place of origin output step according to the present invention.

[0139] Then, after executing step S318, the control unit 130 returns to step S308. Meanwhile, in step S302 described above, if the user does not tap the place of origin estimation button 142a displayed on the menu screen W142, the control unit 130 determines that the place of origin estimation button 142a is not selected and determines "No", and proceeds to step S320.

[0140] In this case, the control unit 130 executes the questionnaire processing program in step S320. Specifically, the control unit 130 (questionnaire processing unit 131) first starts executing the questionnaire processing program shown in Figure 11 in step S500, and in step S502 determines whether or not the questionnaire button 142b displayed on the menu screen W142 has been selected.

[0141] In this case, the control unit 130 (questionnaire processing unit 131) determines that the questionnaire button 142b displayed on the menu screen W142 has been selected and determines "Yes" when tapped, and proceeds to step S504.

[0142] Next, in step S504, the control unit 130 (questionnaire processing unit 131) displays the questionnaire screen on the display screen 141 based on the questionnaire screen information included in the initial information. In this embodiment, the questionnaire screen is a screen for surveying the user's opinion on the estimation accuracy of the estimation results estimated by the place of origin estimation function. In this case, the questionnaire screen displays an answer input field for the user to input answers to the questionnaire, an answer completion button for the user to confirm the answers to the questionnaire, and a back button to transition from the display screen 141 to the menu screen W142.

[0143] Next, in step S506, the control unit 130 (questionnaire processing unit 131) executes the questionnaire screen back subprogram. This questionnaire screen back subprogram is a computer program that returns the content displayed on the display screen 141 from the unanswered questionnaire screen (the currently displayed screen) to the menu screen W142 (the previous screen) according to the user's operation.

[0144] Specifically, the control unit 130 (questionnaire processing unit 131) starts executing the questionnaire screen back subprogram in step S506 and determines whether an operation to return to the menu screen W142 has been performed on the input unit 120. In this case, if the back button displayed on the questionnaire screen is tapped, the control unit 130 (questionnaire processing unit 131) determines "Yes" as an operation to return to the menu screen W142 has been performed and returns to step S502. On the other hand, if the back button is not tapped, the control unit 130 (questionnaire processing unit 131) determines "No" as an operation to return to the menu screen W142 has not been performed and proceeds to step S508. This questionnaire screen back subprogram is always executed while the questionnaire processing program is running.

[0145] Next, in step S508, the control unit 130 (questionnaire processing unit 131) determines whether or not there is input in the answer input field on the questionnaire screen. In this case, the control unit 130 (questionnaire processing unit 131) continues to determine "No" in this determination process until the user starts inputting into the answer input field, and waits for the user to start inputting into the answer input field. Then, when the user starts inputting into the answer input field, the control unit 130 (questionnaire processing unit 131) determines "Yes" in this determination process and proceeds to step S510. In step S510, the control unit 130 (questionnaire processing unit 131) displays the answer content corresponding to the user's input operation in the answer input field.

[0146] Next, in step S512, the control unit 130 (questionnaire processing unit 131) determines whether the response completion button displayed on the display screen 141 has been selected by the user. Here, the response completion button is a button used by the user to notify the control unit 130 (questionnaire processing unit 131) that they have completed the input operation for their answers to the questionnaire.

[0147] In this case, the control unit 130 (questionnaire processing unit 131) continues to determine "No" in this judgment process until the user taps the "finish answer" button displayed on the display screen 141, and returns to step S508. As a result, the control unit 130 (questionnaire processing unit 131) repeatedly executes each of the processes from steps S508 to S512 until the user has finished inputting their answers to the questionnaire. On the other hand, if the user taps the "finish answer" button, the control unit 130 (questionnaire processing unit 131) determines "Yes" in this judgment process and proceeds to step S514.

[0148] Next, in step S514, the control unit 130 (questionnaire processing unit 131) stores the answers to the questionnaire received from the user via the input unit 120 as questionnaire result information in the storage unit 210 via the communication unit 150.

[0149] Then, after executing step S514, the control unit 130 (questionnaire processing unit 131) returns to step S508. Meanwhile, in step S502 described above, if the questionnaire button 142b displayed on the questionnaire screen is not tapped, the control unit 130 (questionnaire processing unit 131) determines that the questionnaire button 142b is not selected and proceeds to step S516. In step S516, the control unit 130 (questionnaire processing unit 131) terminates the execution of the questionnaire processing program and proceeds to step S322 in the function implementation process.

[0150] Next, in step S322, the control unit 130 performs a determination process to determine whether an operation has been performed to transition the display content of the display screen 141 from the menu screen W142 to the initial screen W141. This determination process in step S322 is performed by executing a process similar to that of the menu screen back subprogram in step S212 of the place of origin estimation application. Therefore, the part of the determination process in step S322 that is common with the menu screen back subprogram will be omitted from the explanation as appropriate.

[0151] In this case, if a portion of the initial screen W141 displayed to the right of the menu screen W142 on the display screen 141 is tapped, the control unit 130 determines "Yes" and returns to step S302. On the other hand, if no portion of the initial screen W141 is tapped, the control unit 130 determines "No" in this determination process and proceeds to step S324, ending the execution of the function implementation process.

[0152] Then, after executing step S324, the control unit 130 returns to step S204 in the place of origin estimation application. This place of origin estimation application is executed repeatedly until the user executes the termination subprogram in step S206.

[0153] In this case, after the user has finished interacting with the subject, the user executes the termination subprogram by swiping their finger over the content displayed on the display screen 141 of the output unit 140 (the displayed image resulting from the execution of the place of origin estimation app). As a result, the control unit 130 terminates the execution of the place of origin estimation system 100 in step S216. The user then stops the operation of the user terminal 101 (powers it off) by turning off the power of the tablet terminal device.

[0154] As can be understood from the above description of operation, according to the above embodiment, the place of origin estimation system 100 is configured to include a place of origin estimation means that estimates the place of origin of a subject based on name information representing the subject's name, which includes at least the subject's name. Therefore, the place of origin of a subject can be easily estimated based on readily available name information.

[0155] Furthermore, the present invention is not limited to the embodiments described above, and various modifications are possible as long as they do not depart from the purpose of the present invention. In the description of each modified example, the same reference numerals are used for parts that are the same as in the embodiments described above, and redundant descriptions are omitted.

[0156] For example, in the above embodiment, the place of origin estimation system 100 is configured to include a place of origin estimation means that estimates the place of origin of a subject based on both given name information and non-given name information, which are obtained by separating the subject's given name information. With this configuration, the place of origin estimation system 100 can improve the accuracy of the estimation by estimating the place of origin of a subject based on given name information representing the subject's given name and non-given name information representing the subject's surname and / or third name.

[0157] However, depending on the subject's place of origin, their entire name (full name) may consist only of their given name. Therefore, the place of origin estimation system 100 only needs to be configured to include a means for estimating the subject's place of origin based at least on their given name information. This allows the place of origin estimation system 100 to handle the estimation of places of origin based on various names, such as when a subject's name consists only of their given name. Furthermore, since the place of origin estimation system 100 can estimate a subject's place of origin based at least on their given name information, it can more easily estimate a subject's place of origin based on their given name information, which is easier to obtain than information on their entire name (full name).

[0158] Furthermore, in the above embodiment, the place of origin estimation system 100 is configured to accept input from the user via the input unit 120 for the name information and non-name information of the subject, in separate input fields. This allows the place of origin estimation system 100 to clearly identify the name information and non-name information of the subject based on the information input by the user. However, the place of origin estimation system 100 can also be configured to accept input from the user via the input unit 120 in a state where the name information and non-name information of the subject are not separated.

[0159] Specifically, the place of origin estimation system 100 can be configured to accept input of the subject's given name and ungiven name information from the user via the input unit 120 in the same input field. In this case, the place of origin estimation system 100 may be configured to require the user to input the given name and ungiven name information in a format that makes them distinguishable, or it may be configured to accept the given name information in a format that makes it impossible to distinguish between the given and ungiven name information. Here, a format that makes the given and ungiven name information distinguishable may include, for example, a format in which either the given name or the ungiven name information is enclosed in parentheses, or a format in which symbols such as a space (space), middle dot, slash hyphen, or double hyphen are written between the given and ungiven name information.

[0160] In this case, the place of origin estimation system 100 may be configured to separate the name information of a subject based on the format in which the name information of the subject received via the input unit 120 is entered, if the format is such that the given name information and non-given name information can be distinguished. Alternatively, if the name information of a subject is entered in a format in which the given name information and non-given name information cannot be distinguished, the place of origin estimation system 100 may be configured to estimate the constituent elements of the subject's name (specifically, given name, third name, or surname) based on the name information and separate the name information accordingly.

[0161] Furthermore, in the above embodiment, the place of origin estimation system 100 is configured to accept input of the subject's given name information from the user in the given name input field 143a, and input of non-given name information in the non-given name input field 143b. However, the place of origin estimation system 100 can be configured to accept input of the subject's given name and third name in the given name input field 143a, and input of the subject's surname in the non-given name input field 143b. In this case, the place of origin estimation system 100 can clearly identify the subject's surname information based on the information input by the user. In this case, the place of origin estimation system 100 can consist of information including the subject's given name and / or third name, and information representing the subject's surname.

[0162] Furthermore, in the above embodiment, the place of origin estimation system 100 has a name input field 143A composed of a name input field 143a for inputting the subject's given name information and a non-name input field 143b for inputting the subject's non-name information. With this configuration, the place of origin estimation system 100 clearly distinguishes between the area for inputting the subject's given name (name input field 143a) and the area for inputting the subject's non-name information (non-name input field 143b), making it easier to input the subject's given name information. However, the place of origin estimation system 100 can also have a name input field 143A composed of a field for inputting the subject's given name information, a field for inputting the subject's third name, and a field for inputting the subject's surname. With this configuration, the place of origin estimation system 100 can clearly identify the subject's given name, third name, and surname based on the information input by the user.

[0163] Furthermore, in the above embodiment, the place of origin estimation system 100 is configured with a name information input means (input unit 120) provided by a touch panel on a tablet terminal device. However, the name information input means is not limited to the above embodiment, and can be configured, for example, in place of or in addition to the touch panel in the above embodiment, with a stationary touch panel, keyboard, mouse, camera, image scanner, or microphone.

[0164] In this case, if the name information input means is configured as a camera or image scanner, the subject's name can be extracted as text information from the captured image using OCR (Optical Character Recognition / Reader) and input into the place of origin estimation system 100. Alternatively, if the name information input means is configured as a microphone, the subject's name can be input into the place of origin estimation system 100 using speech recognition technology (SR).

[0165] Furthermore, in the above embodiment, the place of origin estimation system 100 is configured with a name information input means in an input unit 120 integrated into the user terminal 101. However, the name information input means can be any device capable of inputting the name information of a subject to the place of origin estimation system 100, and may not be integrated with the user terminal 101. For example, the name information input means can be configured as a storage medium such as an SD card, USB memory, or voice recorder that stores the name information of the subject, or as an external input device or storage device connected to the place of origin estimation system 100 via a network NW.

[0166] Furthermore, in the above embodiment, the place of origin estimation system 100 is configured with a touch panel display on a tablet terminal device as the place of origin output means (display screen 141). However, the place of origin output means is not limited to the above embodiment, and can be configured with a stationary display or a speaker, for example, instead of or in addition to the touch panel display in the above embodiment. In this case, the speaker includes a device that allows sound to be heard by placing it against the ear, covering it, or inserting it (for example, earphones or headphones).

[0167] Furthermore, the aforementioned microphone or speaker does not necessarily have to be a device that transmits sound using air propagation; for example, it may be a device that transmits sound by vibration of the head or neck, such as bone conduction or skin conduction. In this case, the place of origin estimation system 100 can accept input of the subject's name information by voice or output the estimation result by voice, even if the user's hearing ability is insufficient or the place of use is a noisy environment.

[0168] Furthermore, in the above embodiment, the place of origin estimation system 100 is configured with a display screen 141 integrated into the user terminal 101 as the place of origin output means. However, the place of origin output means can be any device capable of outputting the estimated place of origin, and may not be integrated with the user terminal 101. For example, the place of origin output means can be configured as a storage medium such as an SD card, USB memory, or voice recorder that can store the name information of the subject, or as an external output device or storage device connected to the place of origin estimation system 100 via a network NW.

[0169] Furthermore, in the above embodiment, the place of origin estimation system 100 is configured such that the control unit 230 (input sentence generation unit 231) inserts the subject's given name information and non-given name information, respectively, into the additional sentence generation information to generate the additional sentence. However, the place of origin estimation system 100 may also be configured such that the control unit 230 (input sentence generation unit 231) inserts the subject's given name information, non-given name information, and full name information representing the entire name of the subject, generated by combining the given and non-given name information, into the additional sentence generation information to generate the additional sentence.

[0170] Specifically, the control unit 230 (input sentence generation unit 231) retrieves the name information of subject D, "Mariel", the non-name information of subject D, "Raymond Cruz", and the additional sentence generation information, "The subject's name is <x1>That's it. In this name, the given part <x2>That's it. The parts other than the given name (for example, surname or middle name) are... <x3>It reads out "It is." Next, the control unit 230 (input sentence generation unit 231) inserts the name information "Mariel" and non-name information "Raymond Cruz" of the subject D into the "X2" and "X3" parts of the added sentence generation information, respectively.

[0171] Furthermore, the control unit 230 (input sentence generation unit 231) combines the given name information "Mariel" and the non-given name information "Raymond Cruz" of subject D to generate full name information representing the entire name of subject D in the format "non-given name information, given name information". Next, the control unit 230 (input sentence generation unit 231) inserts the generated full name information into the "X1" portion of the additional sentence generation information to generate additional sentence information.

[0172] Through these processes, the control unit 230 (input sentence generation unit 231) can generate the additional sentence information "The name is <Raymond Cruz, Mariel>. The given name part of this name is <Mariel>. The other parts (for example, surname or middle name) are <Raymond Cruz>.". The control unit 230 (input sentence generation unit 231) can also generate the main sentence, i.e., the input sentence, by inserting this generated additional sentence information into the main sentence generation information.

[0173] Furthermore, in the above embodiment, the input statement consisted of a main statement that indicated an instruction to the generating AI server 300 or, in addition to the instruction, constraints on the response output by the generating AI server 300, and an additional statement that indicated the subject's name information in the input statement. However, the input statement only needs to include the subject's name information and an instruction to the generating AI server 300 to estimate the subject's place of origin based on the subject's name information. Therefore, the input statement can be constructed as a single sentence containing the instruction to the generating AI server 300, constraints, and the subject's name information, without separating the content of the input statement into a main statement and an additional statement.

[0174] Furthermore, in the above embodiment, the input statement is configured to include constraints that limit the format of the response output by the place of origin estimation model. According to this configuration, the place of origin estimation system 100 can improve the accuracy of the response by restricting the output format of the response to the place of origin estimation model using the constraints, thereby allowing the model to perform estimation processing within a specific range. However, the input statement can be configured without these constraints.

[0175] Furthermore, in the above embodiment, the input text was composed of instructions for the place of origin estimation model written in sentence form. However, the input text is not limited to the above embodiment and only requires information that indicates an instruction to estimate the place of origin of the subject in a manner that can convey instructions to the place of origin estimation model. For example, the input text can also consist of text information such as bullet points or tables, audio information, or image information.

[0176] Furthermore, in the above embodiment, the place of origin estimation system 100 is configured to include an input sentence generation unit 231 that generates an input sentence using the subject's name information and input sentence generation information. However, the place of origin estimation system 100 may also be configured so that the input sentence generation unit 231 generates an input sentence using input sentence generation information generated by the place of origin estimation model based on survey result information such as correct / incorrect information, comment information, or feedback information. According to this, the place of origin estimation system 100 can estimate the subject's place of origin based on an input sentence that reflects the user's evaluation of the estimation result (survey result information), and the response output by the place of origin estimation model can be brought closer to what the user expects.

[0177] Furthermore, in the above embodiment, the place of origin estimation system 100 is configured so that the place of origin estimation model estimates and outputs one place of origin that has the highest probability of being the subject's place of origin based on the subject's name information. However, the place of origin estimation system 100 may be configured so that the place of origin estimation model estimates multiple places of origin based on the subject's name information. In this case, the place of origin estimation system 100 can present the user with multiple candidate places of origin for the subject. In this case, the place of origin estimation system 100 may be configured to display the multiple candidate places estimated as the subject's place of origin on the display screen 141 in order from the candidate place of origin with the highest probability of being the subject's place of origin.

[0178] Furthermore, in the above embodiment, the place of origin estimation system 100 was instructed to output only the place of origin of the subject based on the subject's name information. This allows the place of origin estimation system 100 to efficiently perform the estimation process by specializing the place of origin estimation model for estimating the subject's place of origin. However, the place of origin estimation system 100 can also be configured to instruct the place of origin estimation model to estimate the subject's place of origin and related region information based on the subject's name information.

[0179] According to this, the place of origin estimation system 100 can train its place of origin estimation model to learn the relationship between a person's place of origin and region-related information, thereby improving the accuracy of the place of origin estimation model's responses. Furthermore, because the place of origin estimation system 100 has the place of origin estimation model estimate the person's place of origin and region-related information, it can obtain a consistent response that takes into account the relationship between the person's place of origin and region-related information, and can quickly present the estimation results to the user with a single input.

[0180] Furthermore, in the above embodiment, the place of origin estimation system 100 is configured to include language estimation means that estimate multiple languages ​​as languages ​​that the subject can understand. According to this configuration, the place of origin estimation system 100 can present the user with multiple languages ​​as languages ​​that the subject can understand. However, the place of origin estimation system 100 only needs to be configured to estimate at least one language as a language that the subject can understand.

[0181] Furthermore, in the above embodiment, the place of origin estimation system 100 is configured to include a presentation information generation unit 233 that estimates the subject's regional information (in this embodiment, understandable language information and reception information) based on the subject's name information by selecting estimated regional information corresponding to the estimated place of origin from place of origin related information. With this, the place of origin estimation system 100 can easily estimate the subject's place of origin, the subject's understandable language, and reception information based on easily obtainable name information.

[0182] However, the place of origin estimation system 100 only needs to be configured to include at least a place of origin estimation means that estimates the place of origin of the subject based on the subject's name information. Therefore, the place of origin estimation system 100 can be configured so that the presentation information generation unit 233 generates only estimation result screen information for displaying the estimated place of origin on the display screen 141 using the input screen information and response information. In other words, the place of origin estimation system 100 can be configured by omitting the language estimation means, estimated language output means, reception information estimation means, and reception information output means, respectively.

[0183] Furthermore, in the above embodiment, the place of origin estimation system 100 estimated the place of origin of subject D based on the informal name information "Mariel" and informal name information "Raymond Cruz," which represent the name of subject D in katakana, based on the name recognized by the user. According to this, the place of origin estimation system 100 can estimate the place of origin of subject D based on the informal name information of subject D even if the formal name "Mariel Raymundo Cruz" of subject D is unknown.

[0184] Furthermore, even if the user cannot accurately hear or read the name of subject D, the place of origin estimation system 100 can estimate subject D's place of origin from name information based on the pronunciation of subject D's name as recognized by the user. However, the place of origin estimation system 100 can be configured to accept the subject's formal name as name information entered by the subject via the input unit 120, and to estimate the subject's place of origin based on the subject's formal name.

[0185] Furthermore, in the above embodiment, the place of origin estimation system 100 is configured with a large-scale language model for its place of origin estimation model. This allows the place of origin estimation system 100 to estimate a person's place of origin based on the structure of their name and the meaning of the words used in that name, thereby improving the accuracy of the estimation. However, the place of origin estimation system 100 can also be configured to estimate a person's place of origin based on their name information without using a machine learning model.

[0186] Specifically, the place of origin estimation system 100 can estimate the place of origin of a subject using a database that associates the names of multiple subjects with candidate locations that are candidates for each subject's place of origin. This database can consist of information where one candidate location is associated with each subject's name, and / or information where multiple candidate locations are associated with each subject's name. In this case, the subject's name only needs to include at least the subject's name. A computer device equipped with or connected to such a database can be configured to select (estimate) and output one or more candidate locations from the database based on name information entered by the user.

[0187] Furthermore, in the above embodiment, the place of origin estimation system 100 was configured with a large-scale language model for its place of origin estimation model. However, the type of place of origin estimation model is not limited to a large-scale language model. Therefore, the place of origin estimation system 100 can be configured to estimate a person's place of origin based on their name information using a machine learning model other than a large-scale language model.

[0188] Specifically, machine learning models applicable to estimating a subject's place of origin include neural networks (NNs) such as decision tree models, random forests, support vector machines (SVMs), deep neural networks (DNNs), convolutional neural networks (CNNs), recurrent neural networks (RNNs), or long short-term memory (LSTMs), or machine learning algorithms such as k-nearest neighbors, which are configured to estimate a subject's place of origin based on their name information.

[0189] According to this, the place of origin estimation system 100 can handle the estimation of place of origin based on a wider variety of names compared to the case where the database that associates the names of subjects with candidate places of origin of the subjects is used. Here, a wide variety of names include, for example, names not included in the database, special names used within small groups such as limited ethnic groups, regions, or religions, or names that possess characteristics of multiple regions.

[0190] In this case, the place of origin estimation system 100 can estimate the place of origin of the subject based on nominal and non-nominal information using these machine learning models. Here, the estimation method based on nominal and non-nominal information could be, for example, a method of estimating the place of origin of the subject based on two pieces of information, nominal and non-nominal information, or a method of estimating the place of origin of the subject based on each of the nominal and non-nominal information and then using the place of origin that overlaps in each estimation result as the estimated place of origin.

[0191] Here, the place of origin estimation system 100 may be configured to output separate estimation results for each type of information (named information or non-named information) used to estimate the place of origin when estimating the place of origin based on nominal information and non-named information, respectively. Specifically, the place of origin estimation system 100 can output separate estimation results for nominal information and non-named information as the estimation result for subject D, such as "If based on nominal information, possible places of origin are Europe, the United States, the Philippines, or Brazil. If based on non-named information, possible places of origin are the Philippines or Europe."

[0192] Furthermore, estimation methods based on nominal and non-nominal information may include methods that use multiple estimated places of origin based on nominal information as a population and then estimate the estimated place of origin from among them based on non-nominal information, or methods that use multiple estimated places of origin based on non-nominal information as a population and then estimate the estimated place of origin from among them based on nominal information.

[0193] Furthermore, in the above embodiment, the place of origin estimation system 100 is configured to include an update processing unit 234 that performs fine tuning of the place of origin estimation model using survey result information (correct / incorrect information and feedback information) as additional data information. Generally, information on the internet or in books is based on information from before the time the internet content was created or the book was written. On the other hand, since the place of origin estimation system 100 uses survey result information to fine-tune the place of origin estimation model, it can reflect the relationship between (latest) names and the place of origin of the person with that name in the current situation in the place of origin estimation model.

[0194] However, the place of origin estimation system 100 may be configured to perform fine tuning of the place of origin estimation model using information other than the survey results as additional data. Information other than the survey results may include, for example, information on variations in notation where the same person's name is written in various ways, information ranking popular names in each place of origin for specific categories, or information correlating immigration trends to specific regions with changes in immigrant names. Here, information on variations in notation refers to, for example, information where the same person's name is written in different spellings, characters, or written expressions. Specific categories may include, for example, age group, educational level (differences in educational attainment, etc.), occupation, or religion.

[0195] Furthermore, in the above embodiment, the place of origin estimation system 100 is configured to include a process in which the update processing program inputs additional data information to the place of origin estimation model and performs fine tuning of the place of origin estimation model. With this configuration, the place of origin estimation system 100 can repeatedly train the place of origin estimation model by inputting additional data information, and can continuously maintain or improve the accuracy of the answers output by the place of origin estimation model. However, the place of origin estimation system 100 can be configured by omitting the fine tuning execution process in the update processing program.

[0196] Furthermore, in the above embodiment, the place of origin estimation system 100 is configured to include a process in which the update processing program updates the place of origin related information using the place of origin related information included in the feedback information. This allows the place of origin estimation system 100 to update the place of origin related information used to generate estimated region related information with new information, thereby continuously maintaining or improving the accuracy of the estimated region related information. However, the place of origin estimation system 100 can be configured by omitting the process of updating the place of origin related information in the update processing program. That is, the place of origin estimation system 100 can be configured by omitting the update processing program consisting of the fine-tuning execution process and the process of updating the place of origin related information, and can also be configured by omitting the update processing unit 234.

[0197] Furthermore, in the above embodiment, the place of origin estimation system 100 is configured to include a survey processing unit 131 that conducts a survey to investigate the user's opinion on the estimation results estimated by the place of origin estimation function. With this configuration, the place of origin estimation system 100 can provide a survey to investigate the user's opinion on the estimation results estimated by the place of origin estimation function as a convenient function within the place of origin estimation application. However, the place of origin estimation system 100 can be configured without the survey processing unit 131 and the survey processing program that controls the operation of the survey processing unit 131.

[0198] Furthermore, in the above embodiment, the place of origin estimation system 100 is configured by storing the place of origin estimation model in the generating AI server 300. However, the place of origin estimation system 100 can also be configured by storing all or part of the place of origin estimation model in the storage unit 110 of the user terminal 101 or the storage unit 210 of the AP server 200.

[0199] Furthermore, in the above embodiment, the place of origin estimation system 100 is configured with a storage unit 210 in the AP server 200. However, the place of origin estimation system 100 can be configured by omitting the storage unit 210 in the AP server 200 and instead using a storage device connected to the AP server 200 via a network NW. The same applies to the storage unit 110 that constitutes the user terminal 101.

[0200] Furthermore, in the above embodiment, the place of origin estimation program was configured as a single program. However, the place of origin estimation program can also be configured by combining multiple programs. Also, in the above embodiment, the place of origin estimation program was configured to be executed only by the control unit 230 in the AP server 200. However, the device that executes the place of origin estimation program is not limited to the above embodiment, and it can also be configured to be executed by multiple control units.

[0201] Furthermore, in the above embodiment, the place of origin estimation application is configured to include a termination subprogram (step S206). However, if the place of origin estimation application is equipped with separate operating means (e.g., a switch) that receives instructions from the user terminal 101 to terminate the execution of the place of origin estimation application, this termination subprogram (step S206) can be omitted.

[0202] Furthermore, in the above embodiment, the place of origin estimation application is configured to include a menu screen back subprogram (step S212), an input screen back subprogram (step S306), and a questionnaire screen back subprogram (step S506). However, if the user terminal 101 is equipped with a separate operating means (for example, a switch) for receiving instructions to return the display screen 141 to the previous screen, the place of origin estimation application can be configured by omitting these menu screen back subprograms (step S212), input screen back subprograms (step S306), and questionnaire screen back subprograms (step S506). [Explanation of symbols]

[0203] A-D...Target audience, NW...Network, W141...Initial screen, W142...Menu screen, W143...Input screen, W144...Estimated result screen, 100... Place of origin estimation system, 101... User terminal, 110... Memory unit, 120... Input unit, 130...Control unit, 131...Questionnaire processing unit, 140...Output unit, 141...Display screen, 141a...Menu button, 142a...Estimated place of origin button, 142b...Questionnaire button, 143A...Name input field, 143a...First name input field, 143b...Non-first name input field, 143e...Back button, 143B...Estimated result display field, 143c...Estimated place of origin field, 143d... Estimated regional information section, 143da... Language section, 143db... Reception information section, 150...Communication unit, 200...AP server, 210...Storage unit, 230...Control unit, 231...Input text generation unit, 232...Response acquisition unit, 233...Presentation information generation unit, 234...Update processing unit, 250...Communication unit, 300...Generation AI server.

Claims

1. A system for estimating information about the place of origin of a subject, A name information input means for inputting name information representing the name of the subject, which consists of at least the name of the subject, A place of origin estimation means for estimating the place of origin of the subject based on the name information entered via the name information input means, A place of origin estimation system characterized by comprising a place of origin output means that outputs the place of origin estimated by the place of origin estimation means.

2. In the origin estimation system described in claim 1, The aforementioned name information is, The aforementioned name is a character representation of the name, A system for estimating the place of origin, characterized in that the information represents the aforementioned name using a notation different from the official spelling of the name.

3. In the origin estimation system described in claim 2, The aforementioned name information is, A place of origin estimation system characterized in that it is information that represents the name in writing based on the pronunciation of the name recognized by the user using the place of origin estimation system.

4. In the origin estimation system described in claim 1, The aforementioned name information is, Name information representing the name portion of the subject in the aforementioned name, It consists of non-name information representing the part of the name other than the name of the subject. The aforementioned means for estimating the place of origin is, A place of origin estimation system characterized by estimating the place of origin based on the name information and non-name information entered via the name information input means.

5. In the origin estimation system described in claim 1, The aforementioned means for estimating the place of origin is, A system for estimating the place of origin, characterized by using a machine learning model that has learned the relationship between the name of the subject and the place of origin of that subject to estimate the place of origin.

6. In the origin estimation system described in claim 5, The aforementioned machine learning model, A system for estimating place of origin, characterized by being a large-scale language model.

7. In the origin estimation system described in claim 1, further, A language estimation means that estimates the language the subject can understand based on the place of origin estimated by the place of origin estimation means, A place of origin estimation system characterized by comprising: a language estimation means and an estimated language output means that outputs the language estimated by the language estimation means.

8. In the origin estimation system described in claim 7, The language estimation means is A system for estimating the place of origin of the subject, characterized by estimating multiple languages ​​as languages ​​that the subject can understand.

9. In the origin estimation system described in claim 1, further, A reception information estimation means that estimates information useful when receiving the subject based on the place of origin estimated by the place of origin estimation means, A place of origin estimation system characterized by comprising: an attendance information output means that outputs the information estimated by the attendance information estimation means.

10. A program for estimating a place of origin, which causes a computer device to perform a process to estimate information about the place of origin of a subject, The aforementioned computer device, A name information input step for receiving input of name information representing the name of the subject, which consists of at least the name of the subject, A place of origin estimation step which estimates the place of origin of the subject based on the name information entered via the name information input step, A place of origin estimation program characterized by performing a place of origin output step that outputs the place of origin estimated by the place of origin estimation step.

11. A method for estimating a person's place of origin using a place of origin estimation system that estimates information about the person's place of origin, The aforementioned place of origin estimation system, It consists of a place of origin estimation system described in any one of claims 1 to 9, A name information input step of inputting the name information into the name information input means, A place of origin estimation step in which the place of origin estimation means estimates the place of origin of the subject based on the name information entered into the name information input means, A method for estimating a place of origin, characterized by including a place of origin output step which causes the place of origin estimated by the place of origin estimation means to be output to the place of origin output means.

Images