JP6549063B2 - Determination apparatus and determination method - Google Patents

Description

  The present invention relates to a determination apparatus and determination method for determining the strength of an authentication item used for user authentication.

  Currently, in online services, it is common to use a password for user authentication. Such a password is often required to have a predetermined number of characters or more (for example, 6 characters or more) in combination of alphanumeric characters in order to ensure security (see, for example, Non-Patent Document 1). Also, the password including information related to the user, such as the user's phone number and home postal code, can be easily guessed by a third party, so that the service provider should not include the information related to the user in the password. Often forced / recommended.

  Also, in recent years, techniques for checking the strength of passwords have also been developed (see, for example, non-patent documents 2 to 4).

  In order to ensure the security of the password, it is desirable to make the password longer or change it frequently. However, setting an extremely long password or changing the password frequently is cumbersome for the user, and there is a risk of forgetting the password. Therefore, in consideration of the trade-off between convenience and the risk of including the information related to the user in the password, part of the password includes a string related to itself (a string that is related to itself), and It is assumed that many users use the password for a long time without changing it.

  In fact, the most common password composition is "a random alphanumeric combination" followed by "the user's own or family's name" and "the user's own or family's birthday date" There is also a report (see Non-Patent Document 5).

  One method of dealing with password forgetting is to use password reminders. In this method, the user is made to register secret information that only the user himself / herself knows as a password reminder. When the user forgets the password, a password reminder is input instead of the password, and if it is verified that the user is the user, the password is displayed or another password is reissued.

"Login password setting (usage guide)", [online], [search on April 4, 2016], Internet <URL: http://www.mizuhobank.co.jp/direct/goriyo/loginout/log_pass .html> “Password Strength Checker (zxcvbn)”, [online], [search on April 4, 2016], Internet <URL: https://www.agilegroup.co.jp/technote/zxcvbn.html> “Realistic password strength estimation”, [online], [search on April 4, 2016], Internet <URL: https://blogs.dropbox.com/tech/2012/04/zxcvbn-realistic-password-strength- estimation /> "JavaScript (registered trademark) password strength judgment library" Password Checker "", [search on April 4, 2016], Internet <URL: http://www.websec-room.com/passswordchecker> Independent Administrative Institution Information Processing Agency "A Survey on Actual Situation of Online Person Authentication Method", [April 4, 2016 Search], Internet <URL: http://www.ipa.go.jp/files/000040778.pdf >

  Due to the change of the user's attribute, the information related to the user increases year by year. For example, the employee number at the job change destination when the user changes jobs, a new surname by marriage, a new home postal code by relocation, a new mobile phone number by contract change of a mobile phone, the birth of the child by birth of a child There are various things such as day, account number by opening a new account, credit card number by acquiring a credit card, MAC (Media Access Control) address of PC (personal computer) used by the user. Therefore, due to the change of the attribute of the user, an authentication item used for user authentication such as a password or a password reminder may include information related to the user after the fact, which may make it easy for a third party to guess. In addition to the attributes of the user, changes in the external situation may make it easier for a third party to guess the authentication item after the fact.

  Non-patent documents 2 to 4 disclose techniques for checking the strength of passwords, but in these techniques, changes in the posterior strength of authentication items due to changes in user attributes and external conditions are considered. Absent.

  SUMMARY OF THE INVENTION The object of the present invention made in view of the above problems is to provide a determination apparatus and a determination method capable of maintaining the strength of an authentication item even when there is a change in user attributes or external conditions. It is to do.

  In order to solve the above-mentioned subject, a judgment device concerning the present invention is a judgment device which judges intensity of an authentication item used for user's authentication, and it is a storage part which memorizes a character string in which use to the authentication item is undesirable. And a control unit that determines the strength of the authentication item based on the character string stored in the storage unit, the control unit changing at least one of the attribute of the user and the external situation A character string that is not desirable for use for the newly generated authentication item is stored in the storage unit.

  Further, in order to solve the above problems, the determination method according to the present invention is a determination method in a determination apparatus that determines the strength of an authentication item used for user authentication, and the storage unit uses the authentication item. Storing an undesirable character string, and the control unit stores, in the storage unit, a character string that is undesirable for use in the authentication item newly generated with a change in at least one of the user's attribute and the external situation. And determining the strength of the authentication item based on the character string stored in the storage unit.

  According to the determination apparatus and the determination method according to the present invention, it is possible to maintain the strength of the authentication item even when there is a change in the user's attribute or the external situation.

It is a figure which shows the structural example of the system containing the computer as a determination apparatus which concerns on one Embodiment of this invention. It is a figure for demonstrating storing of the character string to NG character string list | wrist shown in FIG. It is a figure which shows the structural example of a NG character string list. It is a figure which shows an example of addition of the derivation | leading-out character string to NG string list. It is a figure for demonstrating an example of the calculation method of a weighting coefficient. It is a figure which shows an example of update of the NG character string list according to the change of the external condition. It is a figure which shows an example of update of the NG character string list according to the change of a user's attribute. It is a flowchart which shows an example of operation | movement of the computer shown in FIG. It is a figure which shows an example of NG character string list. It is a figure which shows an example of NG character string list. It is a figure which shows an example of NG character string list. It is a flowchart which shows another example of operation | movement of the computer shown in FIG. It is a flowchart which shows an example of calculation operation of the similarity degree by CPU shown in FIG. It is a figure which shows another example of NG character string list. It is a flowchart which shows another example of operation | movement of the computer shown in FIG.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. The present invention is not limited to the following embodiments, and it goes without saying that various modifications can be made without departing from the spirit of the present invention.

  FIG. 1 is a view showing an example of the configuration of a system including a computer 10 as a determination apparatus according to an embodiment of the present invention.

  The system shown in FIG. 1 includes a computer 10, an input device 20, and an output device 30. The computer 10 is connected to an input device 20 through which an operation input is performed by the user and an output device 30 for notifying the user etc., and performs predetermined processing in response to the operation input by the user to the input device 20 The result of the process is output to the output device 30. The computer 10 is used, for example, to allow a user to use an online service. In addition, the computer 10 determines the strength of the authentication item used to authenticate the user, such as a password or password reminder, and recommends or forces the change of the authentication item as necessary. Examples of the input device 20 include a keyboard and a mouse. Examples of the output device 30 include a display device and an audio output device.

  Next, the configuration of the computer 10 will be described. The configurations of the input device 20 and the output device 30 are well known to those skilled in the art and are not directly related to the present invention, and thus the description thereof is omitted.

  A computer 10 illustrated in FIG. 1 includes a communication IF (interface) 11, a non-volatile memory 12, and a central processing unit (CPU) 13. The non-volatile memory 12 is an example of a storage unit, and the CPU 13 is an example of a control unit.

  The communication IF 11 communicates with the input device 20 and the output device 30, and outputs, for example, a signal output from the input device 20 to the CPU 13 in response to an operation input by the user. Also, the communication IF 11 outputs the processing result of the CPU 13 according to the user's operation input to the output device 30.

  The non-volatile memory 12 is a storage device that stores various information. For example, the non-volatile memory 12 stores a program for operating the CPU 13 described later. The non-volatile memory 12 also stores authentication items such as a password and a password reminder used for user authentication in the use of the computer 10 and the online service. In addition, the non-volatile memory 12 stores an NG character string list in which character strings that are not desirable for use in an authentication item are stored. In addition, the non-volatile memory 12 may store information such as electronic mail and an address book.

  The CPU 13 reads and operates a program stored in the nonvolatile memory 12 and controls the overall operation of the computer 10. For example, in response to an operation input to the input device 20 by the user, the CPU 13 is a danger based on a character string that is not suitable for use in an authentication item and easiness of matching (similarity by third parties, ease of matching). It stores it in the NG string list in correspondence with the degree (the degree of risk of including the character string in the authentication item). Further, the CPU 13 determines the strength of the authentication item stored in the non-volatile memory 12 based on the NG character string list, and recommends / forces the change of the authentication item as necessary.

  Next, storage of a character string in the NG character string list by the CPU 13 will be described.

  For example, as shown in FIG. 2, when information related to the user such as a birthday and a telephone number is input to the input device 20 by the user as shown in FIG. 2, a character string corresponding to the input birthday and telephone number etc. Is stored in the NG string list as an NG string. Further, when the e-mail and the address book are stored in the non-volatile memory 12, the CPU 13 searches these to extract character strings related to the user such as e-mail address and telephone number, and NG as NG character string Store in string list. In addition, the CPU 13 stores, for example, the MAC address of the computer 10 stored in a network interface card (NIC) for connecting the computer 10 to a network as an NG character string in the NG character string list. Thus, the CPU 13 stores character strings related to users stored in various media of the computer 10 in the NG character string list as NG character strings.

  Next, the configuration of the NG character string list will be described with reference to FIG. 3A.

  As shown in FIG. 3A, the NG character string list includes an item number, an NG character string, a risk score, a weighting coefficient, a risk score after correction (risk), and remarks.

  The item number is a number for identifying an NG character string.

  The NG character string is a character string that is not preferable for use in the authentication item, and is a character string corresponding to information related to the user (information related to the user) such as the user's surname, first name, date of birth, for example. .

  The risk score indicates that an attribute (surname, first name, date of birth, etc.) is given to the NG character string (when the NG character string is regarded as a character string having a meaning corresponding to a certain attribute, not as a simple character string) The numerical value of the risk when using the NG character string for the authentication item is quantified. For example, in the example shown in FIG. 3A, when an attribute of “surname” is added to the NG character string “Yamamoto” of item number “001” (when the NG character string “Yamamoto” is regarded as “surname”), NG When the character string "Yamamoto" is used as an authentication item, the risk score is evaluated as "50".

  The weighting factor is a factor for correcting the risk score by weighting according to the third party analogy of the NG character string and the ease of matching (ease of matching). The weighting factor is determined, for example, by the length (number of characters) of the NG string. Also, the weighting factor is determined, for example, in accordance with the ranking of commonly used surnames when the NG character string indicates a surname. An example of calculation of the weighting factor will be described later.

  The risk score after correction is a value obtained by correcting the risk score using a weighting factor, and indicates the risk based on the easiness of matching of the NG character string. The corrected risk score is calculated by multiplying the risk score by a weighting factor. Therefore, for example, the risk score after correction of the NG character string “Yamamoto” of the item number “001” is “75 (= 50 × 1.5)”.

  The remark indicates an attribute or the like given to the NG character string.

  It should be noted that even if a string derived from an NG string (a derived string), such as changing a part of the NG string, is used as an authentication item, it may be analogized based on the original string. high. Therefore, the derived character string may be stored as the NG character string in the NG character string list. FIG. 3B is a diagram showing an example of the NG character string list in which the derived character string is also stored as the NG character string.

  As shown in FIG. 3B, an NG character string "Yamamoto" of item number "001a" and an NG character string "yamamoto" of item number "001b" are derived from an NG character string "Yamamoto" of item number "001". And are stored.

  The NG character string "YAMAMOTO" is obtained by capitalizing the original NG character string "Yamamoto". Also, the NG character string "yamamoto" is the original NG character string "Yamamoto" written in lower case.

  In the example shown in FIG. 3B, the NG character strings "YAMAMOTO" and "yamamoto" are each assigned the attribute "surname", and the risk score is evaluated as "50" like the original NG character string "Yamamoto". doing. Also, the NG character strings "YAMAMOTO" and "yamamoto" are respectively composed of only upper case letters or lower case letters. Therefore, from the user's point of view, there is no need to perform a Caps operation to switch between upper-case and lower-case input when inputting a character string, which increases convenience. On the other hand, since the NG character strings "YAMAMOTO" and "yamamoto" are respectively composed of only upper case letters and lower case letters, there is no need to mix upper and lower case letters from a third party who tries to infer an authentication item. Becomes easy. Therefore, as shown in FIG. 3B, the weighting factor of the NG character strings "YAMAMOTO" and "yamamoto" is "1.6", which is larger than the weighting factor "1.5" of the NG character string "Yamamoto". . Therefore, the danger score after correction of the NG character strings "YAMAMOTO" and "yamamoto" is "80", which is larger than the danger score "75" after correction of the NG character string "Yamamoto".

  Further, in the example shown in FIG. 3B, the NG character string "19631120" of the item number "003" is derived, and the NG character string "1120" of the item number "003a" and the NG character string "of the item number" 003b " 1963 NOV 1120 ", an NG character string" S 381 120 "of item number" 003 c ", and an NG character string" 20111963 "of item number" 003 d "are stored.

  The NG character string "1120" is the original NG character string "19631120" from which the character string "1963" corresponding to the Christian era is deleted, that is, "birth date" is shortened to "month and day" only. . The NG character string "1963 NOV 1120" is obtained by adding the English abbreviation (NOV) of "Month (November)" in the original NG character string "19631 120". The NG character string "S381120" is a Japanese calendar notation (S38 (Showa 38)) of "year (1963)" of the original NG character string "19631120". The NG character string “20111963” is a European representation of the original NG character string “19631120”.

  In the example shown in FIG. 3B, for example, the NG character strings “1120”, “1963 NOV 1120”, “S381120”, and “20111963”, which are derived from the NG character string “19631120”, are assigned the original NG character string “ It is evaluated as "100" same as 19631120 ".

  On the other hand, for example, the NG character string “1120” is a shortened version of the NG character string “19631120” and is easier to guess. Therefore, the weighting coefficient of the NG character string “1120” is “1.8”, which is larger than the weighting coefficient “1.5” of the original NG character string “19631120”.

  As described above, in the NG character string list, for each NG character string, a danger score when an attribute is given and a weighting coefficient according to the likelihood of analogy of the NG character string are set, and the danger score is A corrected risk score (a risk based on the ease of matching) corrected by the weighting factor is calculated.

  FIG. 4 is a diagram for explaining an example of a method of calculating weighting coefficients. In FIG. 4, an example of calculating a weighting factor in accordance with the ranking and the number of characters of frequently used surnames will be described. In addition, CPU13 performs calculation of a weighting coefficient, for example.

  In the example shown in FIG. 4, the surnames (ranked in kanji) at the top of the rankings are arranged in order. In addition, Roman letters are written for each surname (kanji notation). Such a ranking can be obtained, for example, from the Internet.

  Then, a weighting factor α based on the ranking is set corresponding to each surname. The weighting factor α is set so that the ranking is higher for the surnames of higher ranks. This is because surnames ranked higher are considered to be more likely to encounter a password list attack.

  In addition, a weighting factor β is set based on the number of characters in roman script corresponding to each surname. The weighting factor β is set to be larger as the surname having a smaller number of roman characters. For example, the weighting coefficient β has a minimum value of 1, and is “1.2” when the number of characters is eight, “1.8” when the number is four, and “2.0” when the number is three. It becomes ".

  Furthermore, in order to calculate a weighting factor in which the number of characters written in Roman characters is added to the appearance frequency of the family name, the product (α × β) of the weighting factor α and the weighting factor β is calculated corresponding to each family name. The product value is given as a weighting factor of the NG string indicating each surname. It should be noted that kanji characters are rarely used to input a password, and it is assumed that a password often contains a character string in which a surname is written. Therefore, as shown in FIG. 4, a risk score with higher accuracy can be calculated by adding the weighting coefficient β according to the number of characters in which the surname is written in Roman characters.

  As described above, the CPU 13 updates the NG character string list according to the change of the user's attribute or the external situation. Below, the update of the NG character string list according to the change of an external condition is demonstrated with reference to FIG. In the following, the case of updating the NG character string list shown in FIG. 3B will be described as an example.

  On websites, etc., strings that are dangerous to use for passwords etc. are disclosed. Such character strings are considered to change over time. The CPU 13 updates the NG character string list according to such a change in the external condition.

  The user searches for a web site where a character string that is dangerous to use for a password or the like is published, and the character string published on the web site is input by the user using the input device 20. The CPU 13 stores the input character string in the NG character string list. In FIG. 5, an NG character string “password” of item number “0004”, an NG character string “abcdefg” of item number “0005”, and an NG character string “qwerty” of item number “0006” are added. There is.

  The CPU 13 sets a degree of risk score for the added NG character string. Since each added character string is published as a dangerous character string on the website, the added NG character string has a higher risk score (risk degree) than the user's surname etc. A score of "100" is set.

  Furthermore, the CPU 13 sets a weighting factor for the added NG character string. Here, the CPU 13 sets a weighting factor based on the number of characters of the NG character string. Specifically, the CPU 13 sets a higher weighting coefficient for an NG character string having a smaller number of characters. Then, the CPU 13 calculates the risk score after correction of the NG character string by multiplying the risk score by the weighting coefficient.

  In the above-mentioned example, although a user extracted and explained the example which extracts the dangerous character string on a website and inputs into input device 20, it is not restricted to this. For example, a search engine may be used to extract dangerous strings published on a web site.

  Next, the update of the NG character string list according to the change of the attribute of the user will be described with reference to FIG. Note that, in FIG. 6, job change is described as an example of the change of the attribute of the user. Further, before the user changes jobs, the NG character string list shown in FIG. 3B is stored, and the case of updating the NG character string list will be described as an example.

  It is assumed that the user changes jobs to “SSN Co., Ltd.”. In this case, for example, the company name of the transferee is input to the input device 20 by the user, and the CPU 13 adds the character string “SSN” that constitutes the company name of the transferee as an NG character string to the NG character string list. That is, as shown in FIG. 6, the CPU 13 adds the NG character string "SSN" of the item number "004" to the NG character string list.

  Furthermore, the CPU 13 sets the derived character string (SSn, SsN, Ssn, sSN, sSn, ssN, ssn) combining the upper case and lower case of the NG character string "SSN" as the NG character string of item numbers 004a to 004g, and the NG character Store in column list. By doing this, the character string "SSN" that has become information (character string) newly associated with the user due to a change in the user's attributes (the user's job change) and its derived character string are added to the NG character string list. Can.

  Although FIG. 6 illustrates an example in which job change is an example of the change in the attribute of the user and the company name of the job change destination is added to the NG character string list as an NG character string, the present invention is not limited thereto. Any attribute may be used as long as the attribute is given by the user's life event. For example, the company corporation number, company number (my number), bank account number, last name after marriage, telephone number after relocation, etc. of the company to which the user belongs may be added to the NG string list.

  Next, the determination operation of the strength of the authentication item (password) by the computer 10 according to the present embodiment will be described with reference to the flowchart shown in FIG. In FIG. 7, the computer 10 is a bank ATM (Automatic Teller Machine), and the password for determining the strength will be described using an example in which the password is a password consisting of four digits.

  The CPU 13 reads out the password to be determined (a password consisting of four digits) stored in the non-volatile memory 12. Note that this password is a password whose vulnerability has been checked using a known technique for checking the vulnerability of the password.

  The CPU 13 sets a value (N) indicating the number of the NG character string stored in the NG character string list to 1 (N = 001) (step S11).

  Next, the CPU 13 calculates the similarity R between the personal identification number to be determined and the Nth NG character string in the NG character string list (step S12). The CPU 13 sets the similarity degree R = 100 when the character string that completely matches the Nth NG character string is included in the password to be determined, and sets the similarity degree R = 0 otherwise. Details of the method of calculating the degree of similarity will be described later.

  Next, the CPU 13 determines whether N is the maximum value (whether N is a number corresponding to the last item number of the NG character string list) (step S13).

  If it is determined that N is not the maximum value (step S13: No), the CPU 13 increments N by 1 (step S14), and returns to the process of step S12.

  When it is determined that N is the maximum value (step S13: Yes), the CPU 13 determines whether there is an NG character string whose similarity to the determination target password is equal to or greater than the threshold Rth (step S13). Step S15). The threshold Rth is given to the CPU 13 in advance.

  When it is determined that there is no NG character string whose similarity to the determination target password is equal to or greater than the threshold Rth (step S15: No), the CPU 13 determines that the determination target password has a sufficient strength. (It is determined that the use of the password to be determined is permitted), and the process is ended.

  When it is determined that there is an NG character string whose similarity to the determination target password is equal to or greater than the threshold Rth (step S15: Yes), the CPU 13 determines that the password of the determination target completely matches the NG character string. It is determined that it is necessary to increase the strength of the password to be determined because it is likely to contain a column and the third party may easily guess. In this case, the CPU 13 causes the output device 30 (display device) to display a screen for forcibly changing the password (step S16), and ends the process.

  The operation of the computer 10 described with reference to FIG. 7 will be described in more detail with specific examples.

  In the following, it is assumed that the personal identification number to be determined is “0310”. Further, it is assumed that the NG character string list shown in FIG. 8A is created based on the information related to the user when the bank account is opened (for example, 2007) and stored in the non-volatile memory 12.

  As shown in FIG. 8A, when the bank account is opened, the user's date of birth (monthly day), the user's home postal code (last four digits), the number of the car owned by the user, the user's child (first child) The NG character string corresponding to the information related to the user such as the date of birth (month and day) of is stored. In addition, the risk score, the weighting factor, and the weighting factor after correction are also set and calculated as appropriate.

  Thereafter, it is assumed that the derived string of the stored NG string is added in 2009, and the list is updated to the NG string list as shown in FIG. 8B. That is, it is assumed that a character string in which the user's birth date (month and day) is written in Europe or a character string indicating only the user's birth date is added as an NG character string derived from the user's date of birth. In addition, derived from the date of birth of the first child, a string in which the date of birth (month and day) of the first child is written in Europe, and a string indicating only the age of the first child are added as NG strings. I suppose.

  After that, a second child (second child) is born to the user on March 10, 2015, and information such as the date of birth of the second child is generated according to the change of the user attribute of the birth of the second child, It is assumed that the user inputs to the input device 20. It is assumed that the CPU 13 updates the NG character string list as shown in FIG. 8C according to the input. That is, it is assumed that the date of birth (month and day) of the second child is added as an NG character string. Furthermore, as a derived character string of the second child's date of birth (Month-day), a string in which the second child's date of birth (Month-day) is written in Europe, and a character string indicating only the second child's year of birth Suppose that it is added as a string.

  In the NG character string list (FIG. 8A) at the time of opening the bank account in 2007, there is no NG character string that completely matches the password "0310" to be judged. Therefore, it is determined that the change of the password is unnecessary.

  Also, in the NG character string list (FIG. 8B) after the addition of the 2009 derived character string, there is no NG character string that completely matches the password "0310" to be judged. Therefore, it is determined that the change of the password is unnecessary.

  On the other hand, in the NG character string list after updating according to the birth of the second child (FIG. 8C), the NG character string corresponding to the password "0310" to be judged and the birth date (month and day) of the second child And are completely identical. Therefore, the similarity between the password to be judged and this NG character string is 100. Therefore, the CPU 13 determines that it is necessary to increase the strength of the password, and forces the user to change the password.

  In this way, by updating the NG character string list according to changes in the user's attributes and external conditions, it is possible that the password contains a character string that is dangerous to use for the password, such as information related to the user. If it does, the user can be prompted to change the password. Therefore, the strength of the password can be maintained even when there is a change in the user's attribute or the external situation. In addition, although it demonstrated using the example which updates NG character string list triggered by the birth of 2nd child, it is not restricted to this. For example, the NG string list can be updated in response to various life events such as the user's relocation and the purchase of a car.

  In FIG. 7, when the personal identification number to be determined completely matches the NG character string in the NG character string list, the change of the personal identification number is uniformly requested, but the present invention is not limited to this. For example, even if the password to be judged completely matches the NG character string, the password may have many characters, making it difficult for a third party to guess. In this case, since the risk is considered to be relatively low even if the password to be determined is used, the computer 10 may permit the use of the password to be determined. Hereinafter, the case where the computer 10 performs such an operation will be described with reference to the flowchart shown in FIG. In FIG. 9, the same processing as that of FIG.

  If it is determined that there is an NG character string whose similarity R with the password to be determined is equal to or greater than the threshold Rth (step S15: Yes), the CPU 13 determines that the similarity R is greater than or equal to the threshold Rth (first threshold). It is determined whether or not there is an NG character string whose risk degree score (risk degree score after correction) is equal to or higher than a threshold value Kth (second threshold value) among the NG character strings (step S21). The threshold value Kth is given to the CPU 13 in advance.

  If it is determined that there is an NG character string having a degree of danger score equal to or greater than the threshold Kth (step S21: Yes), the CPU 13 determines that it is necessary to increase the strength of the password to be determined. In this case, the CPU 13 causes the output device 30 to display a screen for forcibly changing the password (step S22), and ends the process.

  If it is determined that there is no NG character string whose risk score is equal to or higher than the threshold Kth (step S21: No), the CPU 13 permits the use of the password to be determined, but the password has a slight risk. It is determined that the password is high, and a screen recommending change of the password is displayed on the output device 30 (step S23), and the process is ended.

  Next, the method of calculating the degree of similarity will be described with reference to the flowchart shown in FIG. In FIG. 10, the authentication item (password) to be determined is “a0sSn + e! 4J”, and as an example, the similarity to the NG character string “sSn” of the item number “004e” of the NG character string list shown in FIG. Description will be made using an example of calculation.

  The CPU 13 compares the password “a0sSn + e! 4J” to be judged with the NG character string “sSn” (step S31). Specifically, the CPU 13 cuts out the character string "a0s" for three characters from the beginning of the password to be judged, and determines whether the NG character string "sSn" and the cut out character string "a0s" completely match. It determines (step S311). Since the NG character string “sSn” and the character string “a0s” do not match one character either, the CPU 13 determines that one of the three characters also does not match.

  Next, the CPU 13 cuts out the character string "0sS" for three characters from the second character of the password to be judged, and determines whether the NG character string "sSn" and the cut out character string "0sS" completely match. It determines (step S312). Since the NG character string “sSn” and the character string “0sS” do not match even one character, the CPU 13 determines that one of the three characters also does not match. Thereafter, the CPU 13 performs the same processing while advancing the position at which the character string for three characters is cut out from the password to be judged one character at a time (steps S313 to S318).

  When the comparison between the last three characters of the password to be judged and the NG character string is completed, the CPU 13 checks whether any character string extracted from the password to be judged completely matches the NG character string It is determined whether or not it is (step S32).

  In the example described above, the NG character string “sSn” and the character string “sSn” for three characters from the third character of the password “a0sSn + e! 4J” to be determined in step S313 also match three of the three characters. Yes, it's a perfect match.

  If it is determined that there is a character string completely extracted from the password to be judged that matches the NG character string (Step S32: Yes), the CPU 13 determines the similarity between the password to be judged and the NG character string. R is calculated to be 100 (step S33).

  If it is determined that none of the character strings cut out from the password to be judged matches the NG character string completely (step S32: No), the CPU 13 determines the similarity between the password to be judged and the NG character string. Calculate R as 0 (step S34).

  In the present embodiment, the similarity R is calculated to be 100 when the NG character string and the character string included in the password to be determined completely match, and the similarity R is 0 otherwise. Although the description has been made using an example of calculating and, it is not limited to this. For example, when the NG character string and the character string included in the password to be determined partially match, an intermediate value between the similarity 100 and the similarity 0 may be calculated as the similarity R.

  As mentioned above, kanji characters are rarely used for passwords. On the other hand, password reminders often accept input of kanji, hiragana and katakana. Below, the case where the intensity | strength of a character string "Yamamoto" is determined as a password reminder is demonstrated as an example.

  When a password reminder is to be determined, as shown in FIG. 11, a character string consisting of kanji, hiragana and katakana is also stored as an NG character string in the NG character string list. In FIG. 11, a character string “Yamamoto” written in kanji is added as an NG character string of the item number “001c” as a derived character string of the NG character string “Yamamoto”. In this case, the danger score of the NG character string “Yamamoto” is set to the same value as that of the NG character string “Yamamoto”. Also, the weighting coefficient of the NG character string "Yamamoto" is calculated based on the surname ranking (see FIG. 4).

  FIG. 12 is a flowchart showing the operation of the computer 10 when the password reminder is to be determined. In FIG. 12, the same processing as that of FIG. 9 is denoted by the same reference numeral, and the description will be omitted.

  The CPU 13 calculates the similarity R between the password reminder to be determined and the Nth NG character string in the NG character string list (step S41). When the similarity with all the NG character strings is calculated up to the NG character string of the last item number of the NG character string list, the CPU 13 determines that the NG character string whose similarity with the password reminder to be determined is the threshold Rth or more It is determined whether it exists (step S42).

  If it is determined that there is an NG character string whose similarity R is equal to or greater than the threshold Rth (step S42: Yes), the CPU 13 determines that there is a risk score (after correction). It is determined whether or not there is an NG character string having a degree of danger score of not less than the threshold value Kth (step S43).

  If it is determined that there is an NG character string whose risk degree score is equal to or higher than the threshold Kth (step S43: Yes), the CPU 13 determines that it is necessary to increase the strength of the password reminder to be determined. In this case, the CPU 13 causes the output device 30 to display a screen for forcing the password reminder to be changed (step S44), and ends the process.

  If it is determined that there is no NG character string having the degree of risk score equal to or higher than the threshold Kth (step S41: No), the CPU 13 permits the use of the password reminder to be determined, but the password reminder is somewhat dangerous. It is determined that the password is high, and a screen recommending change of the password reminder is displayed on the output device 30 (step S45), and the process is ended. Therefore, as in the case of the password, the strength of the password reminder can be maintained even if there is a change in the user's attribute or the external situation.

  As described above, according to the present embodiment, the computer 10 as the determination device that determines the strength of the authentication item includes the non-volatile memory 12 as the storage unit storing the character string that is not preferable for use in the authentication item The CPU 13 as a control unit that determines the strength of the authentication item based on the character string stored in the memory 12, and the CPU 13 newly occurs with a change in at least one of the user's attribute and the external condition, The non-volatile memory 12 stores a string that is undesirable for use in the authentication item.

  Therefore, even if the authentication item becomes information that is related to the user (character string) after the change of the user's attribute or the external situation, the use of the authentication item includes an undesirable character string. The strength of the authentication item can be determined correctly, and the strength of the authentication item can be maintained.

  In the embodiment described above, the example of performing the addition of the NG character string, the determination of the strength of the authentication item based on the stored NG character string, and the like by the computer 10, that is, the terminal local, will be described. However, it is not limited to this. For example, functions such as addition of an NG character string, determination of the strength of an authentication item based on a stored NG character string, and a cloud side connected with a local terminal on the user side via a network like a remote desktop May be provided to a device (such as a server).

  Although the present invention has been described based on the drawings and embodiments, it should be noted that those skilled in the art can easily make various changes or modifications based on the present disclosure. Therefore, it should be noted that these variations or modifications are included in the scope of the present invention.

10 Computer (judgment device)
11 Communication IF
12 Nonvolatile memory (storage unit)
13 CPU (control unit)
20 input devices 30 output devices

Claims (4)

A determination apparatus that determines the strength of an authentication item used for user authentication,
A storage unit that stores a character string that is undesirable for use in the authentication item , and a degree of risk of using the character string based on the ease of matching for the authentication item, corresponding to the character string ;
A control unit that determines the strength of the authentication item based on the character string stored in the storage unit;
The control unit stores, in the storage unit, a character string that is undesirable for use in the authentication item newly generated with a change in at least one of the user's attribute and the external situation .
When a character string whose similarity with the character string indicating the authentication item is equal to or greater than a first threshold is stored in the storage unit and the degree of risk of the character string is equal to or more than a second threshold, the authentication item strength is determined that it is necessary to increase the determination apparatus characterized that you forced to change the authentication item. In the determination apparatus according to claim 1 ,
The control unit stores in the storage unit a character string whose similarity to the character string indicating the authentication item is equal to or greater than the first threshold, and the risk of the character string is less than the second threshold. If it is, the change of the authentication item is recommended to the user. In the determination device according to claim 1 or 2 ,
The determination apparatus according to claim 1, wherein the authentication item is at least one of a password and a password reminder. A determination method in a determination apparatus that determines the strength of an authentication item used for user authentication,
Storing a character string which is not preferable for use in the authentication item , and a degree of risk of using the character string based on the ease of matching in the authentication item, corresponding to the character string ; ,
The control unit stores, in the storage unit, a character string that is undesirable for use in the authentication item newly generated with a change in at least one of the user's attribute and the external condition. Determining the strength of the authentication item based on a string of characters;
A character string whose similarity with the character string indicating the authentication item is equal to or greater than a first threshold is stored in the storage unit in the control unit, and the degree of risk of the character string is equal to or greater than a second threshold Determining that it is necessary to increase the strength of the authentication item, and forcing change of the authentication item .

Images