JP2008158655A - Automatic contract system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an automatic contract system capable of effectively and efficiently eliminating a false contract by efficiently making a membership reference and a credit information reference in consideration of the use frequency of Kana (Japanese syllabary) names without causing an omission in Kana names to be referred to. <P>SOLUTION: A credit determination server 3 carrying out credit determination of a client based on at least attribute information on the client is used to extract Kana names that can give different reading from a Kanji (Chinese character) dictionary or a database 14 for giving different reading of Kanji and storing different reading data of Kanji, for the Kanji name input from the client. The Kana name preferentially used based on the use frequency of Kana used in the past is selected from the extracted Kana names, and the membership or credit information reference is made based on the selected Kana name and the Kana name input from the client. When the person with the Kana name relevant by reference and the client are not the same person, credit determination of the client is carried out. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention includes a credit determination server connected via a communication network to an automatic contract machine that automatically performs a procedure related to a contract such as a loan contract between a financial institution such as consumer finance and a customer. The present invention relates to an automatic contract system, and more particularly to an automatic contract system provided with a technique useful for making a member inquiry or credit information inquiry based on a customer's pseudonym name.

In the financial industry such as consumer finance that provides loans for individuals, members and credit information are inquired based on names, and credit examination is performed based on the results (for example, Patent Document 1). reference). At this time, inquiries are made by using the kana name instead of the kanji name of the customer (see, for example, Non-Patent Document 1). The way of reading (name of pseudonym) may differ. For this reason, if there is a person who wants to make a fake contract because a new contract can be concluded if the reading method (name of pseudonym) is different, not only the name of pseudonym that was reported by the customer when making an inquiry Member inquiry and credit information inquiry are made using possible pseudonym names.
JP 2005-108005 A “STARS (Star Net System)”, [online], i-Net Corporation, [searched on December 14, 2006], 2003, Internet <URL: http://www.i-netc.co.jp/b_gyoumu /b_01_1.html>

  However, when making inquiries using multiple kana names, conventionally, financial institution staff assumed possible kana names and made inquiries based on the kana names. There was a case where the name was missing. If the name of the kana name to be inquired is lost, the member inquiry or credit information inquiry cannot be performed accurately, and a false contract may be missed.

  In addition, since the reading kana of kanji names are diverse, inquiries to Yakumo for possible reading kana (kana names) cannot be made efficiently, and the inquiry costs There were also inconveniences.

  The present invention has been made in view of such circumstances, there is no omission in the name of the kana to be inquired, and the member inquiries and credit information inquiries are efficiently performed in consideration of the use frequency of the reading kana, The main object is to provide an automatic contract system that can effectively and efficiently eliminate fake contracts.

    In order to achieve the above object, an automatic contract system according to the present invention has a credit determination server that acquires customer attribute information via a communication network and performs credit determination of the customer based on at least the attribute information. The credit determination server is configured to receive a customer's attribute information including a Chinese character name inputted from a customer and a kana name corresponding to the Chinese character name, and a Chinese character replacement database storing Chinese character replacement data A kana name extracting means for extracting a readable kana name for the kanji name, and a kana name preferentially used based on the frequency of use of a kana inputted in the past from the kana name extracted by the kana name extracting means. Kana name selection means to be selected, Kana name inputted from the customer, and Kana selected by the Kana name selection means Member inquiry means for making a member inquiry based on a name, the same person determination means for determining whether or not the person of the pseudonym name corresponding to the inquiry by the member inquiry means and the customer are the same person, and And a credit determination unit that performs credit determination of the customer based on at least the customer attribute information when it is determined by the same person determination unit that they are not the same person.

  Therefore, possible kana names extracted for the kanji names entered by the customer using the kanji replacement database are extracted by the kana name extracting means, and further, the kana extracted by the kana name extracting means by the kana name selecting means From the name, a kana name that is used preferentially from the input frequency of the past kana is selected, and the member inquiry is automatically performed based on the selected kana name and the kana name entered by the customer by the member inquiry means. Then, it is determined by the same person determination means whether or not the person with the name of the kana and the customer who was matched by the inquiry of the member inquiry means is the same person, and if it is determined that they are not the same person, the credit determination means Because the customer's credit is judged by the Kana name, the Kana name is selected from the possible Kana names that can be read from the Kanji name in consideration of the input frequency of the past Kana, and the member inquiry is performed. This makes it possible to prevent the omission of queries due to missing pseudonym names that may occur due to empirical judgments, and to reduce unnecessary queries and to make efficient queries. For this reason, it becomes possible to efficiently find member registration of a person who may be the same person as the customer, and to effectively and efficiently eliminate duplicate contracts.

In order to achieve the above object, an automatic contract system according to the present invention has a credit determination server that acquires customer attribute information via a communication network and performs credit determination of the customer based on at least the attribute information. And the credit determination server is configured to
Receiving means for receiving customer attribute information including a kanji name inputted from a customer and a kana name corresponding to the kanji name, and a kana name readable to the kanji name from a kanji replacement database storing kanji replacement data The kana name extracting means for extracting the kana name, the kana name selecting means for preferentially using the kana name extracted based on the frequency of use of the kana entered in the past from the kana name extracted by the kana name extracting means, The name of the pseudonym entered by the customer, the credit information inquiry means for inquiring credit information based on the name of the pseudonym selected by the name of the pseudonym name selection means, The same person determination means for determining whether or not the person and the customer are the same person, and the same person determination means The Most When it is determined, may be constructed and a credit decision means for performing a credit decision of the customer based on at least the attribute information and credit information of the customer.

  Therefore, possible kana names extracted for the kanji names entered by the customer using the kanji replacement database are extracted by the kana name extracting means, and further, the kana extracted by the kana name extracting means by the kana name selecting means From the name, a pseudonym name to be used preferentially is selected from the input frequency of the past kana, and the credit information inquiry means automatically performs the credit information inquiry based on the selected pseudonym name and the pseudonym name entered by the customer. Is called. Then, it is determined by the same person determination means whether or not the person with the name of the pseudonym that was applicable by the inquiry of the credit information inquiry means and the customer are the same person, and if it is determined that they are not the same person, the credit determination Since the credit judgment of the customer is performed by means, among the possible kana names that can be read from the kanji name, the kana name is selected in consideration of the input frequency of the past kana and the credit information inquiry is performed. In addition, it is possible to prevent an omission of a query due to missing a pseudonym name that may occur due to human judgment, and it is possible to reduce a useless query and perform an efficient query. For this reason, it becomes possible to efficiently find member registration of a person who may be the same person as the customer, and to effectively and efficiently eliminate duplicate contracts.

  In any of the above-described configurations, the same person determination unit is configured such that the attribute information other than the customer's pseudonym name and date of birth and the same attribute information of the person with the kana name that was matched by the inquiry match one or more items Alternatively, it may be determined that they are the same person. In addition, the credit determination server, in order to accurately determine whether or not they are the same person, attribute information other than the customer's pseudonym name and date of birth, and the same attribute of the person of the pseudonym name that was applicable by the inquiry A possibility ratio calculating means for calculating a possibility ratio of being the same person based on the number of matching items when the information matches one or more items, and the possibility that the same person determination means is calculated by the possibility ratio calculating means It is good also as a structure which determines whether it is the same person in consideration of a sex ratio.

  Furthermore, in order to make the determination of whether or not they are the same person more accurate, the credit determination server is based on a map information database in which the name of the address is stored. The telephone number of the person of the pseudonym name that was found by the above inquiry based on the telephone information database storing the registered name of the telephone number and the investigation for determining whether the registered name in the name and the customer's kanji name match The survey means having at least one of the surveys for determining whether the registered name in the name and the customer's Chinese character name match or not, and the same person determination unit is the same person in consideration of the survey result by the survey unit It is good also as a structure which determines whether it is.

  In addition, although the above-mentioned kana name extraction means is to extract a readable kana name for a kanji name from a kanji replacement database in which kanji replacement data is stored, instead of or in addition to this, You may make it extract the kana name which can replace the said kanji name from the kanji dictionary.

  As described above, according to the present invention, possible kana names that can be considered for a kanji name input by a customer using a kanji replacement database or kanji dictionary are extracted, and further, the extracted kana names are extracted. Based on the input frequency of the past kana, the kana name is selected, and based on the selected kana name and the kana name entered by the customer, a member inquiry or credit information inquiry is performed. Since the customer's credit is determined when it is determined that the person of the pseudonym name and the customer are not the same person and the person is not the same person, the frequency of use of all possible pseudonym names Inquiries are automatically made with respect to a high name, which prevents an omission of inquiries due to omission of a pseudonym name that may occur due to human judgment, and enables efficient inquiries.

  For this reason, it is possible to make an accurate member inquiry and credit information inquiry to the customer, and it is possible to efficiently eliminate the credit judgment of the customer who may be the same person as the person who is already a member. Therefore, it becomes possible to effectively and efficiently eliminate a fake contract for falsifying a pseudonym name and making a new contract.

  Furthermore, if the attribute information other than the customer's pseudonym name and date of birth and the same attribute information of the person with the kana name that was found in the inquiry match one or more items, there is a possibility that they are the same person depending on the number of matching items. If the ratio is calculated and the possibility ratio is taken into account to determine whether or not they are the same person, whether or not the person with the kana name and the customer who was the result of the inquiry is the same person Judgment can be made accurately, and whether the registered name in the address of the person with the name of the kana name that was applicable by the inquiry matches the name of the customer's kanji or not is applicable. If it is checked whether the registered name in the telephone number of the person with the name of the kana matches the name of the customer's kanji, it is possible to more accurately determine whether or not they are the same person.

  Embodiments of an automatic contract system according to the present invention will be described below with reference to the accompanying drawings.

  In FIG. 1, reference numeral 1 denotes an automatic contract machine for performing processing necessary for concluding a contract between a customer and a company such as a financial institution, and is connected to a credit determination server 3 via a communication network 2 such as the Internet. The various procedures relating to the loan contract and the like are automatically performed by exchanging various information with the credit determination server 3.

  4 is a credit information institution, which is connected to the credit judgment server 3 via the communication network 2 and receives a customer's credit information inquiry message from the credit judgment server 3, and after performing the inquiry process, A service for returning a reply message to the credit determination server 3 is provided.

  Reference numeral 5 denotes an operator terminal, which is connected to the automatic contract machine 1 and the credit determination server 3 via the communication network 2, and monitors the contract processing between the automatic contract machine 1 and the credit determination server 3 or is necessary. In response to this, the customer who uses the automatic contracting machine 1 responds or performs work corresponding to various processing results from the credit determination server 3.

  The automatic contract machine 1 is installed in an unmanned store or the like, and is used to display various information processed by the contract machine and to display the input information and guide the user's operation. For example, the input unit 6 and the display unit 7 are configured by a touch panel display on which the input unit 6 is mounted so as to overlap the display unit 7.

  In addition, the automatic contract machine 1 is a document in which the name of the customer's kanji is written (hereinafter referred to as a kanji name written document, for example, a driver's license for identification or health insurance insured card) Although not shown, a communication control unit that performs data communication processing with the credit determination server 3 via the communication network 2, the scanner 8, the communication control unit, and the input unit 6 is provided. And a control unit configured by a microprocessor that controls the display unit 7 and a storage unit that stores programs and the like.

  On the other hand, the credit determination server 3 is an information processing apparatus having a CPU, a ROM, a RAM, and the like, and uses various customer information received from the automatic contract machine 1 by a predetermined program held in the memory. A credit control is performed, and a communication control unit 11 configured by a router or the like that performs data communication processing with the automatic contracting machine 1 via the communication network 2 and a microprocessor that controls the communication control unit 11 And a control unit 12 including a storage unit storing programs and the like.

  The credit determination server 3 also includes a customer database 13 for storing customer information and credit determination results transmitted from the automatic contract machine 1, a kanji replacement database 14 for storing kanji replacement data, and registered name information for address locations. A map information database 15 for storing the map information including the telephone information database 16 for storing the telephone information including the registered name information of the telephone number is provided.

  As shown in the figure, the control unit 12 of the credit determination server 3 uses the installed software to update the customer attribute information received from the automatic contracting machine 1 based on the Chinese character replacement data stored in the Chinese character replacement database 14. Kana name extraction process that extracts readable kana names for kanji names in the middle, and the kana names that are used preferentially from the past kana input frequency (frequency of use of kana entered in the past) from the extracted kana names Kana selection process for selecting, member inquiry process for inquiring whether or not the user is a member using the customer information stored in the customer database 13 from the selected kana name and the kana name input (declared) by the customer A credit information inquiry process for inquiring credit information to the credit information agency 4 from the selected pseudonym name and the pseudonym name entered (declared) by the customer; Possibility ratio of calculating the probability of being the same person based on the number of matching items between the attribute information other than the kana name and date of birth entered by the customer and the same attribute information of the person with the kana name that was matched by the inquiry By using the registered name information of the address location stored in the map information database 15 in the calculation process, it is determined whether or not the registered name in the address location of the person who corresponds by the inquiry matches the name of the customer's kanji. In addition, using the registered name information of the telephone number stored in the telephone information database 16, it is determined whether or not the registered name in the telephone number of the person who is the result of the inquiry matches the customer's Kanji name. Based on the investigation process to be investigated, the inquiry result, the possibility ratio of being the same person, and the investigation result, the person who was matched by the inquiry and the customer who has processed the automatic contract machine 1 are the same person Same person determination process determines Luke, based on the attribute information and credit information of the customer, having the function of performing such determining credit determination process credit customers.

  FIG. 2 to FIG. 4 show an example of an operation process by the automatic contract machine 1, the credit determination server 3, and the operator terminal 5 as a flowchart. Hereinafter, an example of an operation process of the automatic contract system will be described based on this flowchart. Note that the processing examples shown here include a control program that can be read by the automatic contract machine 1 and the credit determination server 3 via a storage medium such as a ROM chip or a CD-ROM, or a network This is performed by installing the automatic contract machine 1 and the credit determination server 3 by distributing the program via the computer and executing the installed control program.

  First, when the customer operates the input unit 6 of the automatic contract machine 1 and starts the contract procedure, a message prompting consent for confirmation of product contents, use of personal information or the like is displayed on the display unit 7, and confirmation and consent are displayed to the customer. (Step S1).

  In response to this request, if the customer operates the input unit 6 to indicate that he / she does not accept the confirmation and consent, the application is canceled, and thereafter, the contract procedure is not performed and the process is terminated (step S2).

  On the other hand, if the customer operates the input unit 6 and displays an intention to accept the confirmation and consent, the customer is prompted to input the customer's kanji name and its kana name (reading kana). In response, the customer performs a predetermined operation via the input unit 6 to acquire the name of the Chinese character and the name of the kana from the customer (step S3).

  Thereafter, it is required to set a document (kanji name written document) on which the customer's name is written, for example, a driver's license or the like at a predetermined position of the scanner 8, and the customer responds to this request by scanning the document corresponding to the customer. The image of the same document is acquired by setting it at a predetermined position 8 and performing a predetermined operation to perform a reading process (step S4).

  In addition, the customer will be prompted to enter other necessary information necessary for credit review, such as customer attribute information such as date of birth, current address, and telephone number, and necessary listening items necessary for credit such as work and family composition. Other necessary information is acquired (step S5).

  After that, it is determined whether or not the name of the Chinese character is written in the name written document presented by the customer (step S6). This determination is performed by, for example, performing an OCR process on the image of the document written in Chinese character name read in step S4 using an OCR processing program, and automatically determining the presence or absence of the character string of the name.

  In this step S6, when it is determined that the Chinese character name is not described in the document presented by the customer, an error is displayed on the display unit 7 (step S7), and re-reading of the confirmation document of the Chinese character name is requested. The instruction is displayed on the display unit 7 (step S8). Upon receiving this instruction, the customer performs a predetermined operation to read the document written in the name of kanji and re-acquire the image of the document (step S9). Thereafter, in step S6, it is determined again whether or not the Chinese character name is written in the Chinese character name document presented by the customer.

  On the other hand, if it is determined in step S6 that the Chinese character name is described in the document presented by the customer, the Chinese character recognized from the document matches the Chinese character name entered by the customer. It is determined whether or not there is (step S10).

  In this step S10, when it is determined that the kanji recognized from the kanji name document does not match the kanji of the customer's kanji name, it is assumed that there is an error in the kanji name input from the customer. 7 is displayed (step S11), and an instruction for requesting the customer to input the name of the kanji and the name of the kana is displayed on the display unit 7 again (step S12). By doing this, the name of the kanji and the name of the kana are reacquired from the customer (step S13). Thereafter, in step S10, it is determined again whether the kanji recognized from the document matches the kanji of the name entered by the customer.

  On the other hand, if it is determined in step S10 that the kanji recognized from the document matches the kanji name entered by the customer, the customer information obtained from the customer, that is, the kanji name or The pseudonym name (reading pseudonym) and other necessary information are transmitted to the credit judgment server 3 (step S14).

  In the credit determination server 3, when the customer information transmitted from the automatic contract machine 1 is received by the communication control unit 11, the control unit 12 uses the kanji name in the customer information to make an inquiry for each surname and name. Is extracted (step S15).

  Specifically, for the last name, from the replacement data for the last name shown in FIG. 5A stored in the Kanji replacement database 14, all possible replacement data corresponding to the last name of the Kanji name entered by the customer is obtained. Extract. In addition, unlike the last name, the name is stored in a database one by one because it is read differently, and the Kanji entered by the customer from the replacement data for each Kanji of the name shown in FIG. Extract all the replacement data of each kanji corresponding to the name.

  For example, if the name of the Chinese character entered by the customer is “Takeshi Kono”, for the last name, “Kawano” and “Kono”, which are replacement data corresponding to “Kono”, are extracted from the replacement data for the last name in the Kanji replacement database As for the name, “Takeshi”, “Tsushi”, “Gou”, “Takashi”, “Take”, “Tsyo” are the replacement data corresponding to “Takeshi” and “History” from the replacement data related to the name in the Kanji replacement database. , “Hawk”, “shi”, “human”, “fumi” are extracted.

  Then, the usage frequency inputted in the past for the kana name of the name of the Chinese character entered by the customer is confirmed (step S16), and the kana name is selected with priority based on the usage frequency of the kana of this name (step S16). S17).

  Specifically, in the example of “Takeshi” described above, the usage frequency confirmation process in step S16 is combined with other kanji characters for each of the letters “Takeshi” and “History” as shown in FIG. Used in reading (indicated as compound in the figure), reading frequency (number used in the past), and reading in the case of using one character without combining with other kanji (indicated as single in the figure) The usage frequency of the kana (the number of times it has been used in the past) is searched, and the usage ratio of each reading pseudonym is calculated for the composite case and the single case.

  In step S17, based on the past use frequency confirmed in step S16, in the case of compounding, based on the use frequency (rank) when used in compounding each kanji, a combination of reading kana characters is selected. Select by ranking. On the other hand, in the case of a single case, the reading kana is selected with a rank based on the frequency of use (rank) when a single kanji character is used.

  For example, if the kanji in the name part of the name entered by the customer is “Takeshi”, the reading kana with zero or extremely low use frequency (reading kana with a use frequency ratio equal to or lower than a predetermined value) is not adopted. As shown in FIG. 7 (a), the use of the second character's composite reading pseudonym for the first and second use of the combined reading pseudonym (taken, tsuyo). The first and second frequency rankings (Shi, Fumi) are combined, and a kana name is selected along with the ranking (Rank 1 is Takeshi, Rank 2 is Tsuyoshi, Rank 3 is Takefumi, Rank 4 is Tsuyofumi).

  In addition, when the kanji in the name portion of the kanji name entered by the customer is a single character of “Takeshi”, as shown in FIG. The reading kana is selected together with the rank (rank 1 is Takeshi, rank 2 is Yoshi, rank 3 is Go, rank 4 is Takashi).

  Thereafter, in step S18, a combination other than the kana name entered (declared) by the customer from the kana last name extracted in step S15 and the kana name selected in step S17 is determined (step S18). A list of names of kana and surname obtained by combining the kana last name and kana name other than the declaration is displayed on the operator terminal 5 (step S19).

  For example, in the case of the above-mentioned “Takeshi Kono”, the kana last name is “Kawano” and “Kono”, and the kana names are “Takeshi”, “Tsushi”, “Takefumi”, “Tsufumi”, There are 8 combinations (“Takeshi Kawano”, “Yoshi Kawano”, “Takefumi Kawano”, “Yoshi Kawano”, “Takeshi Kouno”, “Yoshi Konotsu”, “Fumi Konotake”, “Fumi Konotake”) Is “Kawa no Tsuyofumi”, seven types of kana names are excluded from the kana names, and the list is displayed on the operator terminal 5.

  After the kana name other than the declaration is confirmed for the kanji name declared by the customer as described above, first, a member inquiry is made using the customer database 13 from the kana name declared by the customer (step S20). Then, it is determined whether or not the contract is applicable based on the name of the declaration kana (step S21).

  If it is determined that the contract does not apply to the filing name, the member inquiry is performed from the extracted kana name other than the filing (the kana name determined in step S18) (step 22), but the contract is made with the filing kana name. If it is determined that there is a match, the attribute information other than the customer's kana name and date of birth (for example, kanji name, address, phone number, work place) It is determined whether one or more items match the attribute information recorded in the customer database 13 (step S23).

  If it is determined in this step S23 that the attribute information of the customer other than the name and date of birth matches with the same attribute information of the corresponding person extracted by the inquiry, there is a possibility that the customer is already a member. Therefore, the fact is displayed on the operator terminal, and confirmation by the operator is performed (step S24).

  On the other hand, when it is determined that the attribute information of the customer other than the name and date of birth of the customer does not match the same attribute information of the corresponding person extracted by the inquiry, other than the declaration is extracted. Member inquiry is made from the kana name using the customer database 13 (step S22), and it is determined whether or not the contract is applicable with the extracted kana name other than the declaration (step S25).

  If it is determined that the contract is not applicable with the extracted pseudonym name other than the declaration, it proceeds to the credit information inquiry after step S32 described later, and it is determined that the contract is applicable with the extracted pseudonym name other than the declaration. In this case, the customer attribute information (for example, kanji name, address, telephone number, work place) other than the pseudonym name and the date of birth is recorded in the customer database 13 of the person who was found by the inquiry. It is determined whether or not the attribute information matches one or more items (step S26).

  If it is determined in step S26 that the customer's attribute information other than the name and date of birth does not match the same attribute information of the person for whom the inquiry has been made, the credit after step S32 to be described later The process proceeds to information inquiry. If it is determined that the customer attribute information other than the pseudonym name and the date of birth matches at least one item with the same attribute information of the person who was found by the inquiry, the customer attribute information and The possibility ratio of the same person is calculated based on the number of items that match the attribute information of the person referred to by the name of kana other than the declaration (step 27).

  Specifically, as shown in FIG. 8 (a), it is determined whether the attribute information other than the kana name and the date of birth, for example, the four attribute information of the kanji name, address, telephone number, and work place match. The possibility ratio is assigned to each item (Kanji name: 25%, address: 30%, telephone number: 30%, work place: 15%), and the same by adding the possibility ratio of the matching items The possibility ratio of the person is calculated, and the result is displayed on the operator terminal 5 together with the ranking of the possibility of the same person as shown in FIG. 8B (step S28).

  For example, if the customer who entered the Kanji name and the person with the extracted Kana name other than the declaration all match the Kanji name, address, phone number, and work place, the possibility ratio of the same person is calculated. 100% (25% + 30% + 30% + 15%), and if the name, address, and phone number match, the probability ratio of the same person is 85% (25% + 30% + 30%) If the name, phone number, and work place match, the possibility ratio of the same person is set to 75% (30% + 30% + 15%).

  Thereafter, the attribute information item of the corresponding person extracted by the inquiry is investigated (step S29). This attribute information item is investigated by extracting the registered name of the address using the map information database 15 from the address of the person with the kana name that was matched by the inquiry, and this registered name is the name of the kanji entered by the customer. The phone number database 16 is used to extract the registered name of the phone number from the phone number of the person with the name of the kana that was matched by the inquiry and the inquiry to determine whether the registered name and the customer At least one of the investigations to determine whether or not the entered kanji name matches is performed.

  Then, based on the results obtained in step S27 and step S29, out of the extracted kana names other than the declaration, the person with the kana name that was the result of the inquiry and the customer who is processing with the automatic contractor 1 Are determined to be the same person (step S30). Specifically, when the possibility ratio of the same person calculated in step S27 is equal to or greater than a predetermined ratio, or as a result of the investigation in step 29, the registered name of the address and telephone number of the person who was found in the inquiry If it is determined that the Kanji name entered by the customer and the Kanji name are the same, the extracted Kana name other than the declaration will be processed by the automatic contracting machine 1 with the person with the Kana name that was the result of the inquiry. It is determined that the customer is the same person, and that fact is displayed on the operator terminal 5 to avoid duplicate contracts, and confirmation by the operator is performed (step S31).

  On the other hand, if it is determined that they are not the same person, the process proceeds to the credit information inquiry after step S32. First, in step S32, credit information is inquired to the credit information organization 4 from the name of the kana declared by the customer. Thereafter, in step S33, credit information is inquired of the credit information organization 4 from the extracted pseudonym name other than the declaration.

  Then, it is determined whether or not there is credit information corresponding to the extracted pseudonym name other than the declaration (step S34), and if there is credit information corresponding to the extracted pseudonym name other than the declaration, the pseudonym name and date of birth are determined. It is determined whether or not the customer attribute information other than the day (for example, kanji name, address, telephone number, office) and the same attribute information of the credit information answer match one or more items (step S35).

  If it is determined in this step S35 that the customer attribute information other than the pseudonym name and date of birth and the attribute information described in the credit information answer match at least one item, the customer attribute information other than the declaration The possibility ratio of the same person is calculated based on the number of items that match the attribute information described in the credit information response obtained with the name of the pseudonym (step S36).

  Specifically, as shown in FIG. 8 described above, it is determined whether the attribute information other than the kana name and the date of birth, for example, the four attribute information of the kanji name, address, telephone number, and work place match. The possibility ratio of the same person is calculated by adding the possibility ratio assigned for each item with the matching attribute information, and the result is displayed on the operator terminal 5 together with the ranking of the possibility of the same person. (Step S37).

  In addition, a survey is performed on the attribute information items of the person with the name of the pseudonym that corresponds to the credit information (step S38). This attribute information item is investigated by extracting the registered name of the address using the map information database 15 from the address described in the credit information response of the person with the name of the pseudonym that was found by the inquiry. The telephone information database 16 is used to make a phone call based on the telephone number described in the credit information response of the person whose kana name is the result of the inquiry and whether or not the Kanji name entered by the customer matches. The registered name of the number is extracted, and at least one of surveys for determining whether or not the registered name matches the name of the Chinese character input by the customer is performed.

  Then, based on the results obtained in step S36 and step S38, out of the extracted kana names other than the declaration, the person of the kana name that was the result of the inquiry and the customer who is processing with the automatic contractor 1 Are determined to be the same person (step S39). Specifically, when the possibility ratio of the same person calculated in step S36 is equal to or greater than a predetermined ratio, and as a result of the investigation of the attribute information item in step 38, If it is determined that the registered name of the telephone number and the name of the Chinese character entered by the customer match, the person with the extracted pseudonym name other than the declaration is the same person as the customer who is processing with the automatic contractor In order to avoid duplicate contracts, this is displayed on the operator terminal and confirmation by the operator is performed (step S40).

  If it is determined in step 34 that there is no credit information corresponding to the extracted kana name other than the declaration, the attribute information of the customer other than the kana name and date of birth is the same attribute information of the credit information answer in step S35. If it is determined that none of the items match, or if it is determined in step S39 that they are not the same person, it can be said that the person is a different person and a new customer. Therefore, a credit decision is made based on the attribute information acquired from the customer and the inquiry result of the credit information for the customer (step S41), and the customer's declaration pseudonym name is recorded in the customer database 13 together with other information (step S42). ), The credit determination result is transmitted to the automatic contract machine 1 to notify the customer (step S43). In response, the credit determination result is displayed on the display unit 7 of the automatic contract machine 1 (step S44).

  Therefore, according to the above-described automatic contract system, the readable syllabary names that can be considered for the Kanji name input from the Kanji replacement database 13 based on the Kanji name input by the customer are extracted, and the past Kana names Kana names that are preferentially used are selected based on the input frequency (frequency of use of the previously input kana), and member inquiry and credit information inquiry based on the selected kana name and the kana name entered by the customer. If there is a match, the same person is determined based on the attribute information other than the pseudonym name and date of birth and the customer's attribute information, and the person is not the same person. If it is determined that the customer's credit is determined, among all the kana names that should be inquired, member inquiries and credit information inquiries will be made to the kana names taking into account the frequency of use. Will be divided, it becomes possible to prevent the inquiry leakage due to omission of kana name to be caused by human judgment, it is possible to efficiently query considering frequency of use of pseudonyms.

  In addition, while taking into account the frequency of use of kana, the member registration of a person who may be the same person but different from the kana name declared by the customer is discovered, and the credit judgment of the customer who may be the same as that person is determined. Since it becomes possible to eliminate, it becomes possible to eliminate a false contract effectively and efficiently.

  Furthermore, if the attribute information other than the customer's pseudonym name and date of birth and the same attribute information of the person with the kana name that was found in the inquiry match one or more items, there is a possibility that they are the same person depending on the number of matching items. A ratio is calculated, and it is determined whether or not they are the same person by taking this possibility ratio into account, and the registered name in the address of the person with the name of the kana that matched the inquiry matches the name of the customer's kanji. And whether or not the registered name in the phone number of the person with the name of the pseudonym that was matched by the inquiry matches the name of the customer's kanji is not. Thus, it is possible to more accurately determine whether the person with the full name and the customer are the same person.

  In the configuration described above, the calculation of the possibility ratio of whether or not they are the same person is not limited to the above-described method, and at least one piece of attribute information is at the stage of moving to step S27 or step S36. Since this is the case, the initial value of the probability ratio is set to 50%, and the ratio when each item matches is set as shown in FIG. 9A, as shown in FIG. 9B. , If the Kanji name, address, phone number, and workplace all match, the customer who entered the Kanji name and the person with the extracted Kana name other than the declaration will have the same probability ratio of being the same person If the possibility ratio of the same person is 100% (50% + 15% + 15% + 15% + 5%), and the three names of Kanji name, address, and telephone number match, the possibility ratio of the same person is 95. % (50% + 15% + 15% + 15 ), And if the name, phone number, and place of work match, the probability ratio of the same person will be 85% (50% + 15% + 15% + 5%). You may make it determine with it being a person on the same day.

  Further, in the above-described configuration, an example in which a kana name that can be read for a kanji name input by a customer is extracted from the kanji replacement database 14 in which kanji replacement data is stored is shown. In order to improve the accuracy, in addition to this, a kanji dictionary database 17 containing kanji dictionary data is provided, or a kanji dictionary function is provided in the kanji replacement database 14, and the kanji name entered by the customer from these kanji dictionaries. It is also possible to extract a pseudonym name that can be read.

  Furthermore, in the above-described configuration, the contract procedure between the automatic contract machine and the credit determination server has been mainly described. However, customer information is transmitted to the credit determination server from a customer terminal connected to the Internet or a terminal installed in the store. A similar process may be performed.

FIG. 1 is a block diagram showing a configuration example of an automatic contract system according to the present invention. FIG. 2 is a flowchart showing an example of operation processing of the automatic contract system according to the present invention. FIG. 3 is a flowchart showing an example of operation processing of the automatic contract system according to the present invention and shows a continuation of FIG. FIG. 4 is a flowchart showing an example of operation processing of the automatic contract system according to the present invention, and shows a continuation of FIG. 5A and 5B are diagrams showing the configuration of the Kanji replacement database. FIG. 5A is a diagram showing the configuration of replacement data relating to the last name, and FIG. 5B is the diagram showing the configuration of replacement data relating to the name. FIG. 6 is a table showing the kana reading kana that composes the name of the kanji name, the use frequency of the kana, the ratio of the use frequency, and the rank of the use frequency. FIG. 7 is a table for explaining a method for selecting a reading name of a name of a kanji name together with a frequency of use. FIG. 7 (a) shows an example of a kanji name consisting of a plurality of characters, and FIG. 7 (b) shows a kanji character. Here is an example of a single name. FIG. 8 is a diagram for explaining a method for calculating the possibility ratio of being the same person. (A) shows the addition ratio for each item for calculating the possibility ratio, and (b) is based on the addition ratio. It is a table | surface which shows the example which calculated the possibility ratio. FIG. 9 is a diagram for explaining another method of calculating the possibility ratio of being the same person, in which (a) shows the addition ratio for each item for calculating the possibility ratio, and (b) shows the addition ratio. It is a table | surface which shows the example which calculated the possibility ratio based on.

Explanation of symbols

1 Automatic contractor 2 Communication network
3 Credit judgment server 14 Kanji replacement database 15 Map information database 16 Telephone information database 17 Kanji dictionary database

Claims (6)

In an automatic contract system configured to have a credit determination server that acquires customer attribute information via a communication network and performs credit determination of the customer based on at least the attribute information.
The credit judgment server is
Receiving means for receiving customer attribute information including a kanji name input from a customer and a kana name corresponding to the kanji name;
A kana name extracting means for extracting a kana name that can be read for the kanji name from a kanji replacement database storing kanji replacement data;
A kana / name selecting means for selecting a kana / name to be used preferentially based on the use frequency of the kana inputted in the past from the kana / name extracted by the kana / name extracting means;
A member inquiring means for inquiring a member based on the pseudonym name inputted by the customer and the pseudonym name selecting means by the kana name selecting means;
The same person judging means for judging whether or not the person of the pseudonym name and the customer who were matched by the inquiry by the member inquiring means are the same person;
An automatic contract system comprising: credit determining means for determining credit of the customer based on at least the customer attribute information when the same person determining means determines that they are not the same person. In an automatic contract system configured to have a credit determination server that acquires customer attribute information via a communication network and performs credit determination of the customer based on at least the attribute information.
The credit judgment server is
Receiving means for receiving customer attribute information including a kanji name input from a customer and a kana name corresponding to the kanji name;
A kana name extracting means for extracting a kana name that can be read for the kanji name from a kanji replacement database storing kanji replacement data;
A kana / name selecting means for selecting a kana / name to be used preferentially based on the use frequency of the kana inputted in the past from the kana / name extracted by the kana / name extracting means;
A credit information inquiry means for inquiring credit information based on the pseudonym name inputted by the customer and the pseudonym name selected by the pseudonym name selection means;
The same person determination means for determining whether the person of the pseudonym name and the customer who were matched by the inquiry by the credit information inquiry means are the same person;
An automatic contract system comprising: credit determining means for determining credit of the customer based on at least the attribute information and credit information of the customer when the same person determining means determines that they are not the same person. The same person determination means is the same person when the attribute information other than the customer's pseudonym name and date of birth and the same attribute information of the person with the kana name that was matched by the inquiry match one or more items. The automatic contract system according to claim 1 or 2, wherein the automatic contract system is determined. The credit determination server determines whether the attribute information other than the customer's pseudonym name and date of birth and the same attribute information of the person with the kana name corresponding to the inquiry match one item or more, according to the number of matching items. A possibility ratio calculating means for calculating a possibility ratio of being the same person;
The automatic identification according to claim 1 or 2, wherein the same person determination means determines whether or not they are the same person in consideration of the possibility ratio calculated by the possibility ratio calculation means. Contract system. The credit determination server is based on a map information database in which the registered name of the address is stored, and whether or not the registered name in the address of the person with the kana name corresponding to the inquiry matches the kanji name of the customer Whether the registered name in the phone number of the person whose kana name was the result of the inquiry matches the kanji name of the customer based on the survey for determining the registered name and the phone information database storing the registered name of the phone number A survey means having at least one of surveys for determining whether or not
3. The automatic contract system according to claim 1, wherein the same person determination means determines whether or not they are the same person in consideration of a result of investigation by the investigation means. The kana name extracting means extracts a readable kana name for the kanji name from a kanji dictionary instead of, or in addition to, extracting a readable kana name for the kanji name from the kanji replacement database. The automatic contract system according to claim 1 or 2, wherein