JP2007207011A5

JP2007207011A5 -

Info

Publication number: JP2007207011A5
Application number: JP2006025715A
Authority: JP
Filing date: 2006-02-02
Publication date: 2009-07-02
Anticipated expiration: 2026-02-02

Claims

A scoring system for calculating a score that reflects the probability of unauthorized use of a credit card,
Authorization data storage means for storing authorization data received every time a credit card is used;
History data storage means for storing, for each user, history data related to credit card users' past credit card usage;
The item value category of the item included in the authorization data stored in the authorization data storage means, the item included in the authorization data stored in the authorization data storage means, and the history data stored in the history data storage means For the item value categories of items generated from the included items , identify the fraud rate at which fraud appeared in each item category, calculate the logit value from the fraud rate, and calculate each category for each item. A logit value conversion table for storing logit values corresponding to
Generated from items included in authorization data stored in the authorization data storage means, items included in authorization data stored in the authorization data storage means , and items included in history data stored in the history data storage means for item, the contribution to the score of each item calculated by logistic regression analysis, a regression coefficient table that associates and stores a regression coefficient corresponding to each item,
An authorization data receiving means for receiving authorization data relating to a credit card usage request;
History data acquisition means for acquiring history data corresponding to the user who made the use request from the history data storage means;
Generated from the item value included in the authorization data received by the authorization data receiving means, the item included in the authorization data received by the authorization data receiving means, and the item included in the history data acquired by the history data acquisition means from the item value of the item, the category specifying means for specifying a category corresponding to the item value of each item required for the score calculation,
Referring to the logit value conversion table, logit value acquiring means for acquiring logit values corresponding to each category specified by the category specifying means;
With reference to the regression coefficient table, regression coefficient acquisition means for acquiring a regression coefficient corresponding to each item whose category is specified by the category specifying means;
The logit value acquired by the logit value acquisition unit is added to the value obtained by multiplying the regression coefficient acquired by the regression coefficient acquisition unit for each item corresponding to the logit value, and the probability of fraud is reflected from the combined value Score calculating means for calculating the score obtained,
A scoring system comprising:

Reflecting the authorization data accepted by the authorization data accepting means, it comprises history data updating means for updating the history data stored in the history data storage means corresponding to a credit card user relating to the authorization data. The scoring system according to claim 1, wherein:

In the logit value conversion table, for the items generated from the items included in the authorization data stored in the authorization data storage means, specify the fraud appearance ratio in which fraud appeared for each item value category in each item, A logit value is calculated from the fraud appearance ratio, and a logit value corresponding to each category for each item is stored in association with each other,
In the regression coefficient table, for items generated from items included in the authorization data stored in the authorization data storage means, the degree of contribution of each item to the score is calculated by logistic regression analysis, and each item is calculated. Is stored in association with the regression coefficient corresponding to
The scoring according to claim 1, wherein the category specifying unit specifies a category corresponding to an item value for each item generated from items included in the authorization data received by the authorization data receiving unit. system.

In the logit value conversion table, the fraud appearance ratio in which fraud appears for each item value category of the item included in the history data stored in the history data storage unit is specified, and the logit value is calculated from the fraud appearance ratio. Logit values corresponding to each category for each item are stored in association with each other,
In the regression coefficient table, for the items included in the history data stored in the history data storage means, the degree of contribution of each item to the score is calculated by logistic regression analysis, and the regression coefficient corresponding to each item is calculated. Is associated and remembered,
The scoring system according to claim 1, wherein the category specifying unit specifies a category corresponding to an item value for each item included in the history data acquired by the history data acquiring unit .

A scoring method for calculating a score that reflects the probability of unauthorized use of a credit card,
The computer system for calculating the score includes an authorization data storage unit that stores authorization data received every time a credit card is used, and history data relating to past credit card usage of the credit card user. And a history data storage unit for storing each,
The computer system stores the item value category of the item included in the authorization data stored in the authorization data storage unit, the item included in the authorization data stored in the authorization data storage means, and the history data storage means For the item value categories of the items generated from the items included in the recorded history data, identify the fraud rate at which fraud appeared for each category in each item, calculate the logit value from the fraud rate, and calculate the item Storing logit values corresponding to each category in a logit value conversion table for storing the logit values in association with each other;
The computer system includes an item included in the authorization data stored in the authorization data storage unit, an item included in the authorization data stored in the authorization data storage unit, and history data stored in the history data storage unit. For the items generated from the included items, calculating the degree of contribution of each item to the score by logistic regression analysis, and storing the regression coefficient corresponding to each item in a regression coefficient table that is stored in association with each other,
The logit value conversion table and regression coefficient table data are updated,
When calculating a score that reflects the probability of fraudulent credit card usage,
An authorization data receiving step in which the computer system receives authorization data relating to a credit card usage request;
A history data acquisition step in which the computer system acquires history data corresponding to the user who has made the use request from the history data storage unit;
The computer system includes the item value of the item included in the authorization data received in the authorization data reception step, the item included in the authorization data received in the authorization data reception step, and the history data acquired in the history data acquisition step. A category identification step for identifying a category corresponding to an item value for each item necessary for calculating the score from the item values of the items generated from the included items ,
The computer system refers to the logit value conversion table, and acquires a logit value corresponding to each category specified in the category specifying step;
The computer system refers to the regression coefficient table, and acquires a regression coefficient corresponding to each item whose category is specified in the category specifying step; and
The computer system adds a value obtained by multiplying the logit value acquired in the logit value acquisition step by the regression coefficient acquired in the regression coefficient acquisition step for each item corresponding to the logit value. A score calculating step for calculating a score reflecting the probability of
A scoring method comprising:

History data update in which the computer system updates the history data stored in the history data storage unit corresponding to the credit card user of the authorization data, reflecting the authorization data received in the authorization data receiving step 6. The scoring method according to claim 5, further comprising steps.

In the logit value conversion table, for the items generated from the items included in the authorization data stored in the authorization data storage unit, the fraud appearance ratio in which fraud appeared for each item value category in each item, A logit value is calculated from the fraud appearance ratio, and a logit value corresponding to each category for each item is stored in association with each other,
In the regression coefficient table, for items generated from items included in the authorization data stored in the authorization data storage unit, the degree of contribution of each item to the score is calculated by logistic regression analysis, and each item is calculated. Is stored in association with the regression coefficient corresponding to
The scoring according to claim 5 or 6, wherein, in the category specifying step, a category corresponding to an item value for each item generated from items included in the authorization data received in the authorization data receiving step is specified. Method.

In the logit value conversion table, the fraud appearance ratio in which fraud has appeared for each item value category of the item included in the history data stored in the history data storage unit is specified, and the logit value is calculated from the fraud appearance ratio. The logit value corresponding to the category corresponding to each item value for each item is associated and stored,
In the regression coefficient table, for the items included in the history data stored in the history data storage unit, the degree of contribution of each item to the score is calculated by logistic regression analysis, and the regression coefficient corresponding to each item is calculated. Is associated and remembered,
The scoring method according to claim 5 or 6, wherein in the category specifying step, a category corresponding to an item value for each item included in the history data acquired in the history data acquiring step is specified.