KR20220095539A - Method for providing weighting using device fingerprint, recording medium and device for performing the method - Google Patents

Abstract

A weighting method using a device fingerprint comprises the steps of: when a user is logged in, measuring a similarity value for each feature of past device fingerprint information and latest logged-in device fingerprint information; calculating entropy indicating a probability with respect to the identifiability of the device fingerprint information for each feature of the device fingerprint; calculating a total similarity score of the device fingerprint by applying the calculated entropy for each feature to a corresponding similarity value of each feature as a weight; and authenticating the user by comparing the total similarity score of the device fingerprint with a preset threshold value. Accordingly, the entropy of similarity for each feature is calculated, and the threshold value is changed through feature entropy to more accurately identify the user.

Description

Weighting method using device fingerprint, recording medium and apparatus for performing the same

The present invention relates to a weighting method using a device fingerprint, a recording medium and an apparatus for performing the same, and more particularly, to calculate the entropy of an appropriate similarity for each feature of the fingerprint, and to set a threshold value through the feature entropy. It relates to a technology to identify a more accurate user by changing it.

With the recent development of Internet technology, online non-face-to-face financial transactions and administrative services have become available, and websites have used plug-in functions to protect users and strengthen security. However, plug-ins cause inconvenience to users due to overlapping installation, reinstallation, and security problems. have.

However, the no-plug-in policy conflicts with the electronic financial transaction security policy of existing financial companies. The financial sector website collects users' media environment information and detects and blocks abnormal financial transactions through installed software to protect users and ensure the stability of electronic financial transactions.

According to the no-plug-in policy, when the existing installation method is converted to a non-installation type, the information that can be collected and detectable is relatively limited, and the effectiveness of preventing abnormal financial transactions is expected to be weakened. Therefore, financial companies need to study alternative technologies and countermeasures that can maintain the functions of collecting and detecting user terminal information without installing additional software.

Device fingerprinting, which has recently been in the spotlight, is a method of tracking a device based on the user's network, operating system, hardware, browser configuration and setting information, etc. Create a print and use it to identify users.

Among many methods for identifying anonymized users in web-based services, a method using differences in device fingerprinting information has been studied. However, the problem of the similarity calculation method is that even for the same user, the item of the feature, the feature information, and the importance of the feature are different at a time, so it is impossible to derive the exact similarity.

KR 10-1767454 B1 JP 6405343 B2 KR 10-2100346 B1

Accordingly, it is an object of the present invention to provide a weighting method using a device fingerprint.

Another object of the present invention is to provide a recording medium in which a computer program for performing the weighting method using the device fingerprint is recorded.

Another object of the present invention is to provide an apparatus for performing a weighting method using the device fingerprint.

In a weighting method using a device fingerprint according to an embodiment for realizing the object of the present invention, when a user logs in, the similarity of past device fingerprinting information and recently logged in device fingerprinting information for each feature measuring a value; calculating an entropy representing a probability for identification of device fingerprinting information for each feature of device fingerprinting; calculating an overall similarity score of device fingerprinting by applying the calculated entropy for each feature as a weight to the similarity value of each corresponding feature; and comparing the overall similarity score of the device fingerprinting with a preset threshold to authenticate the user.

In an embodiment of the present invention, in the step of measuring the similarity value for each feature, at least one of cosine similarity, Jaccard similarity, and Levenshtein Distance may be used.

In an embodiment of the present invention, the calculating of the overall similarity score of device fingerprinting may include calculating the calculated entropy for each feature as a mathematical sum of values obtained by multiplying the similarity value of each corresponding feature by a weight.

In an embodiment of the present invention, the calculating of the entropy may be performed based on the minimum and maximum ranges for each feature of the past device fingerprinting information.

In an embodiment of the present invention, the step of authenticating the user may include: allowing the user's access when the similarity score of the device fingerprinting is equal to or greater than a preset threshold; and when the similarity score of the device fingerprinting is smaller than the threshold, blocking access and requesting additional authentication.

In a computer-readable storage medium according to an embodiment for realizing another object of the present invention, a computer program for performing the weighting method using the device fingerprint is recorded.

In an apparatus for weighting using a device fingerprint according to an embodiment for realizing another object of the present invention, when a user logs in, each feature of past device fingerprinting information and recently logged in device fingerprinting information a similarity measuring unit that measures a similarity value for each individual; an entropy calculation unit for calculating an entropy representing a probability for identification of device fingerprinting information for each feature of device fingerprinting; a similarity scoring unit for calculating an overall similarity score of device fingerprinting by applying the calculated entropy for each feature as a weight to the similarity value of each corresponding feature; and a threshold comparison unit for authenticating the user by comparing the overall similarity score of the device fingerprinting with a preset threshold value.

In an embodiment of the present invention, each similarity measuring unit may use at least one of cosine similarity, Jaccard similarity, and Levenshtein distance.

In an embodiment of the present invention, the similarity scoring unit may calculate the calculated entropy for each feature as a mathematical sum of values obtained by multiplying the similarity value of each corresponding feature by a weight.

In an embodiment of the present invention, the entropy calculator may calculate based on the minimum and maximum ranges for each feature of the past device fingerprinting information.

In an embodiment of the present invention, the threshold comparison unit allows a user's access when the similarity score of the device fingerprinting is equal to or greater than a preset threshold value, and when the similarity score of the device fingerprinting is smaller than the threshold value , access blocking and additional authentication may be requested.

According to the weighting method using the device fingerprint as described above, it is possible to score according to the weight of the fingerprint rather than simply whether the features match, and a higher score can be given as fingerprints with high unique characteristics match. Furthermore, in the case of scoring after grading each fingerprint, it is possible to score in consideration of multiple characteristics such as stability and constancy, as well as considering uniqueness.

1 is a block diagram of an apparatus for assigning weights using a device fingerprint according to an embodiment of the present invention.
2 is exemplary data for comparing the similarity of device fingerprinting according to the present invention.
3 shows the results of calculating the similarity according to each similarity measuring method based on the example data of FIG. 2 .
FIG. 4 is a result of a method of confirming whether device fingerprinting information matches with the conventional 1:1 matching in the example data of FIG. 2 .
5 is a result of confirming the similarity in the example data of FIG. 2 by the entropy-based similarity scoring method according to the present invention.
6 is an example of a device fingerprinting user authentication model used in the present invention.
7 is a flowchart of a weighting method using a device fingerprint according to an embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS [0012] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS [0010] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS [0010] Reference is made to the accompanying drawings, which show by way of illustration specific embodiments in which the present invention may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the present invention. It should be understood that the various embodiments of the present invention are different but need not be mutually exclusive. For example, certain shapes, structures, and characteristics described herein with respect to one embodiment may be implemented in other embodiments without departing from the spirit and scope of the invention. In addition, it should be understood that the location or arrangement of individual components within each disclosed embodiment may be changed without departing from the spirit and scope of the present invention. Accordingly, the detailed description set forth below is not intended to be taken in a limiting sense, and the scope of the present invention, if properly described, is limited only by the appended claims, along with all scope equivalents as those claimed. Like reference numerals in the drawings refer to the same or similar functions throughout the various aspects.

Hereinafter, preferred embodiments of the present invention will be described in more detail with reference to the drawings.

1 is a block diagram of an apparatus for assigning weights using a device fingerprint according to an embodiment of the present invention.

The weighting apparatus 10 (hereinafter referred to as the apparatus) using a device fingerprint according to the present invention gives a weight score by considering entropy for each feature in the similarity scoring evaluation for each feature of the device fingerprint.

Referring to FIG. 1 , an apparatus 10 according to the present invention includes a similarity measurement unit 110 , an entropy calculation unit 130 , a similarity scoring unit 150 , and a threshold value comparison unit 170 .

In the apparatus 10 of the present invention, software (application) for performing weighting using a device fingerprint may be installed and executed, and the similarity measuring unit 110, the entropy calculating unit 130, and the similarity scoring unit may be installed and executed. The configuration of the unit 150 and the threshold comparison unit 170 may be controlled by software for weighting using the device fingerprint executed in the apparatus 10 .

The device 10 may be a separate terminal or may be a part of a module of the terminal. In addition, the similarity measuring unit 110 , the entropy calculating unit 130 , the similarity scoring unit 150 , and the threshold value comparing unit 170 may be configured as an integrated module or one or more modules. have. However, on the contrary, each configuration may be formed of a separate module.

The device 10 may be movable or stationary. The apparatus 10 may be in the form of a server or an engine, and may be a device, an application, a terminal, a user equipment (UE), a mobile station (MS), or a wireless device. (wireless device), may be called other terms such as a handheld device (handheld device).

The device 10 may execute or manufacture various software based on an operating system (OS), that is, the system. The operating system is a system program for software to use the hardware of the device, and is a mobile computer operating system such as Android OS, iOS, Windows Mobile OS, Bada OS, Symbian OS, Blackberry OS and Windows series, Linux series, Unix series, It can include all computer operating systems such as MAC, AIX, and HP-UX.

When a user logs in, the similarity measuring unit 110 measures a similarity value for each feature of past device fingerprinting information and recently logged in device fingerprinting information.

A user's PC device and browser each have their own unique values when processing data. Fingerprinting is to use this to identify individuals. Some browsers work effectively in preventing the browser from identifying you while increasing the level of privacy.

In the user authentication model of device fingerprinting, device fingerprinting is generated when the user inputs ID and PW. Device fingerprinting first collected compares IP and IP obtained through WebRCT to determine whether to bypass, saves generated device fingerprinting, or blocks access and requires additional authentication.

In addition, if the user's fingerprint is stored, the past fingerprinting information and the recently logged-in fingerprint are compared, and if the same, access is permitted. On the other hand, if the fingerprinting information is different, the similarity scoring of each fingerprint is compared, and if it is above a threshold, access is allowed. Otherwise, access is blocked and additional authentication is required.

In the present invention, similarity scoring evaluation for each feature considering entropy in device fingerprinting and a weight score for entropy for each feature are given.

In device fingerprinting, information changes depending on time or device, so even if connected from the same device, it may not match exactly. Through this, users cannot access or an inconvenient situation that requires additional authentication occurs, so it is important to properly divide users using entropy.

Similarity scoring methods include cosine similarity, Jaccard similarity, and Levenshtein distance. The problem with the similarity calculation method is that similarity cannot be accurately derived by calculating the similarity for the same number of items or for all fingerprint features. The method below is an example of a similarity method suitable for each feature feature, not a method of comparing the similarity by integrating the features.

Similarity is a measure of how similar two pieces of data are. It is important to measure the similarity between data in all fields, but it is the most basic in the classification field, and through this, more complex things can be learned. Device fingerprinting helps to more accurately classify the similarity calculation method according to the features.

The similarity can be measured for each feature through the following three similarity calculation methods. However, the following similarity calculation methods are only examples, and other similarity calculation methods may be used.

1. Cosine Similarity

The cosine similarity is a method of measuring the similarity through the angle formed by two vectors. If this is expressed as an equation, it is as Equation 1 below.

[Equation 1]

The smaller the angle between the two vectors, the higher the similarity, and the larger the angle, the smaller the similarity. Cosine similarity can be compared through length normalization and is not affected by frequency or word order.

2. Jaccard Similarity

Jacquard similarity is one of the methods of measuring similarity through two sets of words after making two sentences into sets of words. If this is expressed as an equation, it is as Equation 2 below.

[Equation 2]

The measurement method divides the intersection of two sets into the union of two sets. It is mainly used in similar document identification systems and is not affected by the order or quantity of words.

3. Levenshtein Distance

를

로 변환하는 최소 횟수를 두 문자열간의 거리로 정의한다. 이를 수식으로 나타내면 아래의 수학식 3과 같다.Levenstein distance is one string

cast

Defines the minimum number of times to convert to as the distance between two strings. This is expressed as Equation 3 below.

[Equation 3]

It is used to correct spelling errors, search for similar phrases, and judge DNA sequence similarity. The similarity is measured to see if the same value is obtained when the string is changed to a minimum through insertion, deletion, and modification.

2 is an example for comparing the similarity of device fingerprinting according to the present invention, and FIG. 3 is a cosine similarity, Jaccard similarity, and Levenshtein distance based on the example data of FIG. 2 . ) to show the results of calculating the similarity for each.

The entropy calculation unit 130 calculates an entropy representing a probability of identification of device fingerprinting information for each feature of device fingerprinting.

Each feature is diverse, and the more information of many attributes is used, the more identifiable. However, there are various criteria for features as well as identification. It can be selected and used according to the primitive characteristics of the standard. There is a need to rank features by criteria such as distinctiveness (uniqueness), stability, immutability, and privacy.

An indication of the probability that a fingerprint having a specific attribute is unique is called identification. Discrimination is expressed by measuring information entropy. Entropy refers to the average amount of information, and the lower the probability of a specific event, the greater the uncertainty and the greater the entropy, that is, the greater the amount of information. Equation 4 below is an example of calculating entropy, and entropy can be calculated by other methods.

[Equation 4]

Equation 4 is expressed by calculating the minimum-maximum range based on the past features. The value of entropy can be expressed as a grade indicating importance according to the range of features. In addition, entropy is not an absolute value because it may vary depending on the size and format of the collected data. Therefore, accurate user authentication is possible only when weights are assigned according to features.

The similarity scoring unit 150 calculates an overall similarity score of device fingerprinting by applying the calculated entropy for each feature as a weight to the similarity value of each corresponding feature.

The device score is calculated as the sum of the mathematical similarity calculations for each printing attribute. As an example, the fingerprinting grade is determined according to the change of the feature. Grade 1 is Cookie Enabled, HTTP encoding grade 2 is Screen, Timezone, and Grade 3 can be classified as Useragent, Font, and Plugin.

The device score is obtained by multiplying the reference weight by the entropy value through the entropy obtained from each feature. If a feature of the past device fingerprint is matched, 1 point is given and a weight is assigned to it, and a score is calculated as in Equation 5 below.

[Equation 5]

The threshold value comparison unit 170 compares the overall similarity score of the device fingerprinting with a preset threshold value to authenticate the user.

When the similarity score of the device fingerprinting is equal to or greater than a preset threshold, the user's access is permitted. On the other hand, when the similarity score of the device fingerprinting is smaller than the threshold, access blocking and additional authentication may be requested.

The entropy score of each feature calculated in Equation 5 is included in the threshold criterion. A higher score is given as fingerprints with high unique characteristics match, and if it is greater than the threshold, authentication fails if it is less than the authentication success threshold. This is expressed as Equation 6 below.

[Equation 6]

FIG. 4 is a result of a method of confirming whether device fingerprinting information matches or not through conventional 1:1 matching in the example of FIG. 2 . In this case, as a mathematical similarity scoring method, a known similarity calculation method is used according to data characteristics. A value of 0 or 1 is given to each feature according to whether or not they match, and matching is determined only when the number of fingerprints is equal to or greater than a certain number.

5 is a result of confirming the similarity in the example data of FIG. 2 by the entropy-based similarity scoring method according to the present invention. This process is a method that reflects the identification of device fingerprinting, and multiplies the score and entropy, which is a value between 0 and 1, which calculates the similarity of the properties of each device fingerprint, and adds them all together, and was used as the total scoring.

Referring to FIG. 5 , when the entropy-based similarity scoring method according to the present invention is applied, the similarity value of the two strings is calculated by comparing the value stored for User A and the value of New, which is a value collected in the current session. do. In addition, after setting the entropy value of each attribute as a weight, the similarity value and the value calculated by multiplying the weight are all added to use the similarity of New to User A.

Accordingly, in the present invention, it is possible to score according to the weight of the fingerprint, not simply whether it is a match, and a higher score can be given as fingerprints with high intrinsic characteristics match. Furthermore, in the case of scoring after grading each fingerprint, it is possible to score in consideration of multiple characteristics such as stability and constancy, as well as considering uniqueness.

6 is an example of a device fingerprinting user authentication model used in the present invention. 7 is a flowchart of a weighting method using a device fingerprint according to an embodiment of the present invention.

The weighting method using the device fingerprint according to the present embodiment may be performed in substantially the same configuration as that of the apparatus 10 of FIG. 1 . Accordingly, the same components as those of the device 10 of FIG. 1 are given the same reference numerals, and repeated descriptions are omitted.

Also, the weighting method using the device fingerprint according to the present embodiment may be executed by software (application) for performing weighting using the device fingerprint.

In the present invention, a weight score is given in consideration of entropy for each feature in the similarity scoring evaluation for each feature of device fingerprinting.

A user's PC device and browser each have their own unique values when processing data. Fingerprinting is to use this to identify individuals. Some browsers work effectively in preventing the browser from identifying you while increasing the level of privacy.

Referring to FIG. 6 , in the user authentication model of device fingerprinting, when a user accesses and inputs ID and PW (step S00), device fingerprinting is started (step S10). A fingerprint is generated based on information input by the user and information such as CryptoFP (step S20).

In the case of first-collected device fingerprinting (step S30), it is determined whether to bypass the IP by comparing the IP with the IP obtained through WebRCT (step S40). If it is not the bypassed IP, the generated device fingerprint is stored (step S50), and in the case of the bypassed IP, access blocking and additional authentication are requested (step S90).

If the user fingerprint is stored (step S30), the past fingerprinting information is compared with the recently logged-in fingerprinting (step S60), and if they are the same, the user's access is allowed through user authentication (step S80).

On the other hand, if the fingerprinting information is different, the similarity scoring of each fingerprint is compared (step S70), and if it is greater than a threshold, user access is allowed, otherwise access is blocked and additional authentication is requested (step S70).

In the present invention, in step S70, the similarity scoring evaluation for each feature considering the entropy of device fingerprinting and a weight score for the entropy for each feature are given.

In device fingerprinting, information changes depending on time or device, so even if connected from the same device, it may not match exactly. Through this, users cannot access or an inconvenient situation that requires additional authentication occurs, so it is important to properly divide users using entropy.

Referring to FIG. 7 , in the weighting method using a device fingerprint according to the present embodiment, when a user logs in, a similarity value is measured for each feature of past device fingerprinting information and recently logged in device fingerprinting information. (Step S71).

Similarity scoring methods include cosine similarity, Jaccard similarity, and Levenshtein distance. The problem with the similarity calculation method is that similarity cannot be accurately derived by calculating the similarity for the same number of items or for all fingerprint features. The methods presented above are examples of a similarity method suitable for each feature feature, not a method of comparing the degree of similarity integrating features.

Similarity is a measure of how similar two pieces of data are. Measuring the similarity between data in any field is important, but it is the most fundamental in classification and allows you to do more complex things. Device fingerprinting helps to more accurately classify the similarity calculation method according to the features.

For each feature of device fingerprinting, an entropy representing a probability for identification of device fingerprinting information is calculated (step S72).

Each feature is diverse, and the more information of many attributes is used, the more identifiable. However, there are various criteria for features as well as identification. It can be selected and used according to the primitive characteristics of the standard. There is a need to rank features by criteria such as distinctiveness (uniqueness), stability, immutability, and privacy.

An indication of the probability that a fingerprint having a specific attribute is unique is called identification. Discrimination is expressed by measuring information entropy. Entropy refers to the average amount of information, and the lower the probability of a specific event, the greater the uncertainty and the greater the entropy, that is, the greater the amount of information.

For example, it can be calculated based on the minimum and maximum ranges for each feature of the past device fingerprinting information, but any entropy calculation method is applicable.

An overall similarity score of device fingerprinting is calculated by applying the calculated entropy for each feature as a weight to the similarity value of each corresponding feature (step S73).

The device score is calculated as the sum of the mathematical similarity calculations for each printing attribute. As an example, the fingerprinting grade is determined according to the change of the feature. Grade 1 is Cookie Enabled, HTTP encoding grade 2 is Screen, Timezone, and Grade 3 can be classified as Useragent, Font, and Plugin.

The device score is obtained by multiplying the reference weight by the entropy value through the entropy obtained from each feature. In other words, the total scoring can be calculated by multiplying the entropy and the score, which is a value between 0 and 1, which calculates the similarity of features of each device fingerprinting, and adding them all together.

The user is authenticated by comparing the overall similarity score of the device fingerprinting with a preset threshold (step S74).

When the similarity score of the device fingerprinting is equal to or greater than a preset threshold, the user's access is allowed, and when the similarity score of the device fingerprinting is less than the threshold, access blocking and additional authentication may be requested.

According to the present invention, it is possible to score according to the weight of the fingerprint, not whether it is simply a match, and a higher score is given to a fingerprint having a high intrinsic characteristic, so that a user can be identified more accurately.

Such a weighting method using a device fingerprint may be implemented as an application or may be implemented in the form of program instructions that may be executed through various computer components and recorded in a computer-readable recording medium. The computer-readable recording medium may include program instructions, data files, data structures, etc. alone or in combination.

The program instructions recorded in the computer-readable recording medium are specially designed and configured for the present invention, and may be known and available to those skilled in the computer software field.

Examples of the computer-readable recording medium include a hard disk, a magnetic medium such as a floppy disk and a magnetic tape, an optical recording medium such as a CD-ROM and DVD, and a magneto-optical medium such as a floppy disk. media), and hardware devices specially configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like.

Examples of program instructions include not only machine language codes such as those generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like. The hardware device may be configured to operate as one or more software modules for carrying out the processing according to the present invention, and vice versa.

Although the above has been described with reference to the embodiments, those skilled in the art can variously modify and change the present invention within the scope without departing from the spirit and scope of the present invention described in the claims below you will understand

Device fingerprints that can be collected from users using standard web functions vary widely. Identification (uniqueness) can increase as information of many attributes is used to identify users, but indiscriminate collection of information should be avoided. Therefore, it is necessary to analyze according to various criteria such as whether the device fingerprint is sufficiently meaningful information and whether it can be collected from the browser, and it is necessary to select and use core information according to the purpose of use.

Through a web browser that supports web standards, it is possible to analyze the existing device authentication and identification more accurately and precisely through the entropy score by dividing it into 10 high-level information and 50 low-level information such as IP information, browser information, and system information. . As such, device fingerprinting can be utilized in various forms such as user access history management, tracking, blacklisting, and device scoring for security purposes.

10: Weighting device using device fingerprint
110: similarity measurement unit
130: entropy calculation unit
150: similarity scoring unit
170: threshold value comparison unit

Claims (11)

when a user logs in, measuring a similarity value for each feature of past device fingerprinting information and recently logged in device fingerprinting information;
calculating an entropy representing a probability for identification of device fingerprinting information for each feature of device fingerprinting;
calculating an overall similarity score of device fingerprinting by applying the calculated entropy for each feature as a weight to the similarity value of each corresponding feature; and
Comparing the overall similarity score of the device fingerprinting with a preset threshold value, and authenticating the user;
The method of claim 1, wherein measuring the similarity value for each feature comprises:
A weighting method using a device fingerprint, using at least one of cosine similarity, Jaccard similarity, and Levenshtein Distance.
The method of claim 1, wherein calculating the overall similarity score of the device fingerprinting comprises:
A weighting method using a device fingerprint, wherein the calculated entropy for each feature is calculated as a mathematical sum of values obtained by multiplying the similarity value of each corresponding feature by a weight.
The method of claim 1, wherein calculating the entropy comprises:
A weighting method using a device fingerprint, which is calculated based on the minimum and maximum ranges for each feature of past device fingerprinting information.
The method of claim 1, wherein authenticating the user comprises:
allowing a user's access when the similarity score of the device fingerprinting is equal to or greater than a preset threshold; and
When the similarity score of the device fingerprint is less than the threshold, access blocking and requesting additional authentication; Containing, a weighting method using a device fingerprint.
A computer-readable storage medium in which a computer program for performing the weighting method using the device fingerprint according to claim 1 is recorded.
a similarity measuring unit for measuring a similarity value for each feature of past device fingerprinting information and recently logged in device fingerprinting information when a user logs in;
an entropy calculation unit for calculating an entropy representing a probability for identification of device fingerprinting information for each feature of device fingerprinting;
a similarity scoring unit for calculating an overall similarity score of device fingerprinting by applying the calculated entropy for each feature as a weight to the similarity value of each corresponding feature; and
and a threshold value comparison unit for authenticating the user by comparing the overall similarity score of the device fingerprint with a preset threshold value.
The method according to claim 7, wherein each of the similarity measuring units comprises:
A weighting apparatus using a device fingerprint using at least one of cosine similarity, Jaccard similarity, and Levenshtein Distance.
The method of claim 7, wherein the similarity scoring unit,
A weighting apparatus using a device fingerprint for calculating the calculated entropy for each feature as a mathematical sum of values obtained by multiplying the similarity value of each corresponding feature by a weight.
The method of claim 7, wherein the entropy calculation unit,
A weighting apparatus using a device fingerprint that calculates based on the minimum and maximum ranges for each feature of past device fingerprinting information.
The method of claim 7, wherein the threshold value comparison unit,
When the similarity score of the device fingerprinting is greater than or equal to a preset threshold, the user's access is allowed,
When the similarity score of the device fingerprint is less than the threshold value, access blocking and additional authentication are required, a weighting apparatus using a device fingerprint.

Images