User authentication device, user authentication method, and user authentication computer program

Abstract

A user authentication device includes: a collection part collecting information of a user; a generation part generating a question for the user on the basis of the information of the user collected by the collection part and a skill model of the user; a presentation part presenting the question for the user generated by the generation part to the user; a reception part receiving, from the user, a response to the question presented by the presentation part; and a determination part determining authentication of the user on the basis of the response received by the reception part.

Description

BACKGROUND OF THE INVENTION

1. Field of the Invention

The present invention relates to a user authentication device, a user authentication method, and a user authentication computer program.

2. Description of the Related Art

The general problem coped with by a current user authentication method is to authenticate a person with high safety using an application (at the time of log in a computer network, for example).

Two-element authentication is a technique established to confirm the identity of a user. For this practical use, the second authentication method is required to satisfy the followings:

• • The method is safe and constantly available.
  • The method suppresses a burden to the minimum (to be entertaining if possible).

Any of the methods being used today does not satisfy all of the above-described standards. Moreover, it is considerably preferable to authenticate a user on user's channel.

Most of the two-element authentication methods include, as at least one element, a biometric authentication process. For this function, one method trusts the other method. For example, it is difficult to confirm that a photo is taken at the time and that a photo is not a copy (of a screen or a piece of paper as the means conducted by hackers in the past). Every time the authentication method is improved to be safer, a new problem tends to appear.

U.S. patent application No. 2015/0096002 (Method of Criminal Profiling and Person Identification Using Cognitive/Behavioral Biometric Fingerprint Analysis) discloses a use of cognitive/behavioral identification information for identifying a criminal psychological state (see U.S. patent application No. US 2015/0096002). As a solution in this patent literature, a fingerprint at the time of a criminal behavior is formed for a general user (the fingerprint is formed while a user is showing a criminal behavior).

U.S. patent application No. 2013/0185071 (VERIFYING A USER) discloses the determination of a response time by a speech ability using voice transition patterns (see U.S. patent application No. US 2015/0096002). The solution in this patent literature is a solution by biometric voice authentication.

SUMMARY OF THE INVENTION

It is an object of the present invention to at least partially solve the problems in the conventional technology.

However, the above-described conventional techniques do not enable, in two-element authentication, user authentication with higher accuracy. This is because the above-described conventional techniques have the following problems.

First, U.S. patent application No. 2015/0096002 discloses a use of cognitive/behavioral identification information for identifying a criminal physical state, while it does not disclose a solution for identifying a person.

Second, U.S. patent application No. 2013/0185071 discloses the determination of a response time by a speech ability using voice transition patterns, while it does not measure an online response time for a problem requiring a specific skill.

The above and other objects, features, advantages and technical and industrial significance of this invention will be better understood by reading the following detailed description of presently preferred embodiments of the invention, when considered in connection with the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a diagram illustrating an example of a user authentication system according to a first embodiment;

FIG. 2 is a block diagram illustrating a configuration example of a user authentication device according to the first embodiment;

FIG. 3 is a flowchart illustrating an example of a flow of the entire user authentication processing according to the first embodiment;

FIG. 4 is a flowchart illustrating an example of a flow of user authentication determination processing according to the first embodiment; and

FIG. 5 is a diagram illustrating a computer executing a computer program.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

The following will specifically describe the user authentication device, the user authentication method, and the user authentication computer program according to the invention, with reference to the enclosed drawings. Note that the invention is not limited to the embodiment described in the following.

First Embodiment

The following will sequentially describe a configuration of the user authentication system, a configuration of the user authentication device, a concrete example of the user authentication processing, a flow of the entire user authentication processing, and a flow of the user authentication determination processing, and will lastly describe the effects of the embodiment.

Configuration of User Authentication System

The following will specifically describe a configuration of a user authentication system (appropriately referred to as this system) 100 according to the first embodiment, with reference to FIG. 1. FIG. 1 is a diagram illustrating a configuration example of a user authentication system according to the first embodiment. This system 100 includes a user authentication device 10, a user 20, and a user information source 30. Here, the user authentication device 10, the user 20, and the user information source 30 are connected to be communicable wiredly or wirelessly through a given communication network (not illustrated). Note that the user authentication system illustrated in FIG. 1 may include a plurality of user information sources 30.

First, the user authentication device 10 collects information of the user 20 from the user information source 30 (Step S1). Next, in accordance with the collected information of the user 20, the user authentication device 10 constructs a skill model of the user 20 (appropriately referred to as a “user skill model”, a “user model”, or a “model”), and generates a question set for authenticating the user 20 (Step S2).

The above-described Steps S1 and S2 are steps for generating, with the use of the authentication system, a model for each user regarding the ability (appropriately referred to as a “user skill”) of solving a complicated but simple task (appropriately referred to as a “question” or a “problem”). Such a model is constructed first in accordance with the background of the user 20 (for example, a Mathematical Olympiad participant, a musician, or the like). After various authentication problems, the model is constantly refined for each user by a machine learning method. Moreover, the model learns how the user 20 copes with the task.

Furthermore, in the above-described Steps S1 and S2, the user authentication device 10 generates a question/problem for the user 20 so as to test a response time and the characteristics at the time of solving the problem of the user 20, and similarly generates other indices related to the question.

Then, the user authentication device 10 presents the question to the user 20 (Step S3), and receives a response from the user 20 (Step S4). Here, the user authentication device 10 determines the authentication of the user 20 on the basis of the response from the user 20 (Step S5).

In the above-described Step S3 to Step S5, if the response time and the characteristics at the time of solving the problem of the user are within model parameters, the user 20 is authenticated. Otherwise, the user 20 may be tested further or warned. In this manner, when the user 20 intends to log in a network or the like or when the user 20 intends to access specific data or the like with high confidentiality, it is possible to confirm the identity of the user 20. Note that the details of the user authentication determination processing will be described later in “Flow of user authentication determination processing”.

Then, the user authentication device 10 notifies the user 20 of a determination result of the user 20 (Step S6). Furthermore, the user authentication device 10 updates the skill model of the user 20 on the basis of the response from the user 20 (Step S7).

In the above-described Step S7, it is possible to refine, by an unsupervised machine learning method, a model regarding the ability of the user 20 for solving a task within a domain, on the basis of the response time of the user 20, the accuracy of the result, and other parameters.

The above-described user authentication system 100 according to the first embodiment constructs a signature such as a predicted response time required for a person to perform a complicated task, and is a new authentication technique that can be used as the second authentication method of the two-element authentication. For example, unlike the authentication technique of the above-described U.S. patent application No. 2015/0096002, this system 100 can construct a model regarding a user skill, and present a question with a different difficulty level generated by artificial intelligence (AI) in accordance with a specific user so as to authenticate the specific user. Moreover, unlike the authentication technique of the above-described U.S. patent application No. 2013/0185071, this system 100 can perform processing based on user modeling using behavioral analysis and skill-based behaviors. Furthermore, this system 100 authenticates a user on a network, and constructs a skill model unique to each regular user, thus enabling user authentication with higher accuracy.

Configuration of User Authentication Device

The following will specifically describe the configuration of the user authentication device 10 according to the first embodiment with reference to FIG. 2. FIG. 2 is a block diagram illustrating a configuration example of the user authentication device according to the first embodiment. The user authentication device 10 includes an input unit 11, an output unit 12, a communication unit 13, a storage unit 14, and a control unit 15.

The input unit 11 controls input of various kinds of information to the user authentication device 10. The input unit 11 is, for example, a mouse, a keyboard, or the like, and receives input of setting information and the like to the user authentication device 10. Moreover, the output unit 12 controls output of various kinds of information from the user authentication device 10. The output unit 12 is, for example, a display or the like, and outputs setting information and the like stored in the user authentication device 10.

The communication unit 13 controls data communication with other devices. For example, the communication unit 13 performs data communication with communication devices. Moreover, the communication unit 13 can perform data communication with an operator terminal (not illustrated).

The storage unit 14 stores various kinds of information referred to by the control unit 15 for operation and various kinds of information obtained by the control unit 15 during operation. The storage unit 14 includes a user information storage 14a, a skill model storage 14b, and a question set storage 14c. Here, the storage unit 14 is, for example, a random access memory (RAM), a semiconductor memory element such as a flash memory, or a storage device such as a hard disk or an optical disk. Note that although the storage unit 14 is provided in the user authentication device 10 in the example of FIG. 2, the storage unit 14 may be provided outside the user authentication device 10, or a plurality of storage units may be provided.

The user information storage 14a stores information related to a resume of the user 20, a hobby of the user 20, a skill of the user 20, and the like that are collected from the user information source 30. Moreover, the skill model storage 14b stores information related to a skill model constructed for each user. Moreover, the question set storage 14c stores a plurality of questions and correct answers generated for each user.

The control unit 15 controls the entire user authentication device 10. The control unit 15 includes a collection part 15a, a construction part 15b, a generation part 15c, a presentation part 15d, a reception part 15e, a determination part 15f, an application part 15g, and an update part 15h. Here, the control unit 15 is, for example, an electronic circuit such as a central processing unit (CPU) or a micro processing unit (MPU) or an integrated circuit such as an application specific integrated circuit (ASIC) or a field programmable gate array (FPGA).

The collection part 15a collects information of the user 20. For example, the collection part 15a collects information of the user 20 using text analysis, natural language processing, or semantic analysis. The collection part 15a also collects information of the user 20 by analyzing the user information source 30 such as a resume, a LinkedIn, and ResearchGate of the user 20, and a nonprofessional social network such as Facebook including information of hobbies and other skills of the user 20. Meanwhile, the collection part 15a stores the collected information of the user 20 in the user information storage 14a.

The construction part 15b constructs a skill model of the user 20 using the information of the user 20 collected by the collection part 15a. For example, the construction part 15b constructs an initial model regarding a problem-solving skill of the user 20, using all of the information of the user 20 collected by the collection part 15a and a general-purpose model preliminarily constructed by the construction part 15b and then refined. Meanwhile, the construction part 15b stores the information related to the constructed skill model of the user 20 in the skill model storage 14b.

The generation part 15c generates a question for the user 20 on the basis of the information of the user 20 collected by the collection part 15a and the skill model of the user 20. For example, the generation part 15c generates, as a question for the user 20, a numerical question, a mathematic question, a physical question, a mechanical question, a reading exercise, or a speed chess. The generation part 15c also generates a question set for user authentication on the basis of the skill model of the user 20, and predicts time required for the user 20 to answer the questions correctly. Meanwhile, the generation part 15c stores, in the question set storage 14c, the generated question set, and the predicted time required for the user 20 to answer the questions correctly.

The presentation part 15d presents a question for the user 20 generated by the generation part 15c to the user 20. If a response to the question does not satisfy a given condition, the presentation part 15d presents another question to the user 20. For example, if the response time of the user 20 or the accuracy deviates largely from a predicted range, the presentation part 15d further presents one or more different questions to the user 20.

The reception part 15e receives, from the user 20, a response to the question presented by the presentation part 15d. For example, the reception part 15e receives, as a response to a question, an answer to the question, a response time, and the characteristics at the time of solving the question. Meanwhile, the reception part 15e may store the received various kinds of information in the storage unit 14.

The determination part 15f determines authentication of the user 20 on the basis of the response received by the reception part 15e. For example, if the accuracy of an answer to the question, the response time, and the characteristics at the time of solving the question are within a range of parameters set by the skill model of the user 20, the determination part 15f determines that the authentication of the user 20 is successful. If one of the accuracy of an answer to the question, the response time, and the characteristics at the time of solving the question is out of a range of parameters set by the skill model of the user 20, the determination part 15f determines that the authentication of the user 20 is failed.

If a response to the question does not satisfy a given condition, the application part 15g applies a given flag to the response to the question. For example, if the determination part 15f has determined that the authentication of the user 20 is failed, the application part 15g applies a flag indicating that a user is suspicious to the response of the user 20. Moreover, if the response time of the user 20 or the accuracy deviates largely from the predicted range, the application part 15g applies a flag indicating that a problem exists in cognitive ability to the response of the user 20. Meanwhile, the application part 15g may store the information related to the applied flag in the storage unit 14. The application part 15g may also transmit the information related to the applied flag to the user 20 and a system administrator.

The update part 15h updates a skill model of the user 20 on the basis of the response to the question received by the reception part 15e. For example, the update part 15h inputs, as data, the accuracy of an answer to the question, the response time, or the characteristics at the time of solving the question of the user 20, and updates the skill model of the user 20 using an unsupervised machine learning method. The update part 15h may also input the information of the user 20 newly collected by the collection part 15a and update the skill model of the user 20.

Concrete Example of User Authentication Processing

The following will describe concrete examples of the user authentication processing according to the first embodiment. The following will describe the concrete example 1 to the concrete example 3. However, the embodiment does not limit to the following concrete examples.

Concrete Example 1

In the first embodiment, the accuracy of a model regarding the problem-solving ability of each user can be improved constantly. In the following concrete example, there will be described processing of improving the accuracy of a model by increasing the number of dimensions used.

First, in the processing of the concrete example 1, the user authentication device 10 uses a plurality of domains for each user and constructs super space of user ability over such domains. For example, the user authentication device 10 constructs, as constructed super space of user ability, super space of numerical questions, chess problems, inference problems, and the like.

Moreover, in the processing of the concrete example 1, the user authentication device 10 can add the number of dimensions of skills acquired by the user 20 at a workplace. For example, the user authentication device 10 can ask a network system administrator of a company a question about a consequence predicted when a specific network component of a network managed by the administrator is changed. The user authentication device 10 can also generate arithmetic questions with high complicity for an accountant.

As described above, in the processing of the concrete example 1, the user authentication device 10 increases the number of dimensions used and keeps refining, with every authentication question, a model for each user by machine learning. Therefore, even if an attacker solves the same task as the user 20, the attacker needs to solve the task in the same way as the user 20 in the same time frame. For example, a chess problem can be solved in various ways, and the specific way of solving the problem by the user is grasped. Thus, if the way of solving the problem is different, a flag is applied. In this manner, the accuracy of user authentication is improved, thereby making it difficult for an intruder to pretend to be a true user.

Concrete Example 2

In the following concrete example, there will be described processing using a reading exercise as a different task example, which is in the same category as the above-described processing. In the processing of the concrete example 2, the user authentication device 10 generates a sample text, and observes the movement of the eyes of the user 20 reading the text, with a camera provided in the device (laptop or mobile). Here, each person has an extremely different pattern of eye movement during reading a text on a screen, and the pattern may differ also depending on reading ability, reading comprehension, and reading habit of each user.

The user authentication device 10 constructs a model regarding the eye movement of the user 20 during reading a text, improves the model by observing the user 20 reading a text, and adjusts the model by a machine learning method. The user authentication device 10 then uses such a model so as to determine a user to be authenticated on a network. The possibility that a swindler can imitate even the eye movement of the user 20 is low. Thus, with the processing of the concrete example 2, it is possible to protect the system from deepfake being used increasingly by hackers.

The above-described concrete example 1 and concrete example 2 are acting solutions constantly improving access monitoring in a single channel, and are methods considerably difficult for an attacker to break. More importantly, the method is difficult to break by automatic attacks. That is, if a response is excessively quick or slow, a flag is applied as suspicious, and thus simply knowing an answer is not enough.

Concrete Example 3

In the following concrete example, there will be described processing of thoroughly checking the intelligence of a user to be authenticated. For example, in a case of a system administrator or a programmer under the influence of alcohol or drugs, a problem in cognitive ability appears in an answer to a question. In the processing of the concrete example 3, if such a response is received, the user authentication device 10 applies such information as a flag to the response. Therefore, with the processing of the concrete example 3, it is possible to prevent unintentional damages to a network or a product of a company.

Flow of Entire User Authentication Processing

The following will specifically describe a flow of the entire user authentication processing according to the first embodiment with reference to FIG. 3. FIG. 3 is a flowchart illustrating an example of a flow of the entire user authentication processing according to the first embodiment.

First, the collection part 15a of the user authentication device 10 collects information of the user 20 from the user information source 30 (Step S101). Here, the collection part 15a may obtain the information of the user 20 from the user information storage 14a of the storage unit 14.

Next, the construction part 15b constructs a skill model of the user 20 using the information of the user 20 collected by the collection part 15a (Step S102). Here, the construction part 15b may construct the skill model of the user 20 using the information of the user 20 stored in the user information storage 14a. Note that the user authentication device 10 may obtain a skill model of the user 20 constructed by another device.

Subsequently, the generation part 15c generates a question set including questions (problems) for the user 20 and correct answers on the basis of the information of the user 20 collected by the collection part 15a and the skill model of the user 20 (Steps S103). Here, the generation part 15c generates, using the skill model of the user 20, a problem required to be solved within a few seconds by the user 20. However, the time required for response is not especially limited. For example, if a problem is an entertaining game, the generation part 15c may generate a problem required to be solved within a longer time.

Moreover, the generation part 15c predicts a time required for the user 20 to respond to a question (response time) on the basis of the skill model of the user 20 (Step S104). Here, the generation part 15c may predict the characteristics at the time of solving a question of the user 20, on the basis of the skill model of the user 20.

Then, the presentation part 15d presents the question for the user 20 generated by the generation part 15c to the user 20 (Step S105). Here, the presentation part 15d may present the question for the user 20 in a form of text data or image data through the output unit 12, or present the question for the user 20 in a form of voice data through the output unit 12.

Moreover, the reception part 15e receives, from the user 20, a response to the question presented by the presentation part 15d (Step S106). Here, the reception part 15e may receive the response to the question in a form of text data or image data through the input unit 11, or receive the response to the question in a form of voice data through the output unit 11.

Thereafter, the determination part 15f determines authentication of the user 20 on the basis of the response received by the reception part 15e (Step S107). The details of the user authentication determination processing by the determination part 15f will be described later in “Flow of user authentication determination processing”. Moreover, the determination part 15f notifies the user 20 of a determination result of whether the authentication is successful, for example (Step S108). Here, the determination part 15f may notify the user 20 of the determination result through the output unit 12 or transmit the determination result to the terminal of the user 20 or a system administrator through the communication unit 13.

Moreover, if the response to the question does not satisfy a given condition, the application part 15g applies a given flag to the response to the question (Step S109). Here, the application part 15g applies a flag indicating that a user is suspicious, a flag indicating that a problem exists in cognitive ability of the user 20, or the like. However, the kind of a flag is not especially limited.

Lastly, the update part 15h updates the skill model of the user 20 on the basis of the response to the question received by the reception part 15e (Step S110), and finishes processing. Note that the processing at Step S110 may be performed before the processing at Step S107.

Flow of User Authentication Determination Processing

The following will specifically describe a flow of the user authentication determination processing according to the first embodiment with reference to FIG. 4. FIG. 4 is a flowchart illustrating an example of a flow of the user authentication determination processing according to the first embodiment. First, the determination part 15f of the user authentication device 10 extracts an answer to a question, a response time, and characteristics at the time of solving the question, which are included in the response received by the reception part 15e. The determination part 15f performs the following processing of Steps S201 to S203 and determines authentication of the user 20.

First, the determination part 15f determines whether the extracted response time of the user 20 is within a range of a model parameter of an expected range (Step S201). Here, if the response time is within the model parameter (Yes at Step S201), the processing is shifted to Step S202. In contrast, if the response time is not within the model parameter (No at Step S201), the processing is shifted to Step S205.

Secondly, the determination part 15f determines whether the accuracy of the extracted answer result of the user 20 is within a range of a model parameter of an expected range (Step S202). Here, if the accuracy of the answer result is within the model parameter (Yes at Step S202), the processing is shifted to Step S203. In contrast, if the accuracy of the answer result is not within the model parameter (No at Step S202), the processing is shifted to Step S205.

Thirdly, the determination part 15f determines whether the extracted other characteristics (characteristics at the time of solving the problem) of the user 20 are within a range of a model parameter of an expected range (Step S203). Here, if the other characteristics are within the model parameter (Yes at Step S203), the processing is shifted to Step S204. In contrast, if the accuracy of the result is not within the model parameter (No at Step S203), the processing is shifted to Step S205.

If numerical values based on all of the responses are within the model parameters in the processing of Step S201 to S203, the determination part 15f determines that the authentication of the user 20 is successful (Step S204), and finishes processing. Note that the determination part 15f can also perform the processing of Step S201 to S203 in a different order.

In contrast, if any one of numerical values based on all of the responses is not within the model parameter in the processing of Step S201 to S203, the determination part 15f determines whether another question is to be presented to the user 20 (Step S205). If another question is to be presented to the user 20 (Yes at Step S205), the processing is shifted to Step S201 where the determination part 15f performs processing again on an answer to a new question. If another question is not to be presented to the user 20 (No at Step S205), the determination part 15f determines that the authentication of the user 20 is failed (Step S206), and the processing is finished.

Here, if the response time, the accuracy, or other characteristics of the user 20 deviates largely from the model parameter of an expected range, the application part 15g applies a flag indicating that the user 20 is suspicious. Moreover, if the user 20 is suspicious, the presentation part 15d may allow the user to solve one or more problems before taking some actions such as warning. If the user 20 is suspicious, the determination part 15f may also warn the user 20 immediately after the first question. Note that a security system administrator or the like can change settings of the kind of warning issued by the determination part 15f and the situation in which a warning is issued.

Effects of First Embodiment

First, in the user authentication processing according to the above-described first embodiment, the information of a user is collected, and a question for the user is generated on the basis of the collected information of the user and a skill model of the user. The generated question for the user is presented to the user, and a response to the presented question is received from the user. Then, the authentication of the user is determined on the basis of the received response. In this manner, this processing enables, in the two-element authentication, user authentication with higher accuracy.

Secondly, in the above-described user authentication processing according to the first embodiment, the information of the user is collected using text analysis, natural language processing, or semantic analysis. In this manner, more detailed information of the user is collected, whereby this processing enables, in the two-element authentication, user authentication with higher accuracy.

Thirdly, in the above-described user authentication processing according to the first embodiment, a skill model of the user is constructed using the collected information of the user. In this manner, a more effective skill model of the user is constructed, whereby this processing enables, in the two-element authentication, user authentication with higher accuracy.

Fourthly, in the above-described user authentication processing according to the first embodiment, there is generated, as a question for a user, a numerical question, a mathematic question, a physical question, a mechanical question, a reading exercise, or a speed chess. In this manner, a more effective question for a user is generated, whereby this processing enables, in the two-element authentication, user authentication with higher accuracy.

Fifthly, in the above-described user authentication processing according to the first embodiment, if an answer to a question does not satisfy a given condition, another question is presented to the user. In this manner, a plurality of questions are presented to a suspicious user, whereby this processing enables, in the two-element authentication, user authentication with higher accuracy.

Sixthly, in the above-described user authentication processing according to the first embodiment, if the accuracy of an answer to a question, the response time, and the characteristics at the time of solving the question of the user are within a range of parameters set by the skill model of the user, it is determined that the authentication of the user is successful. In this manner, the response of the user is determined more effectively, whereby this processing enables, in the two-element authentication, authentication with higher accuracy.

Seventhly, in the above-described user authentication processing according to the first embodiment, if a response to a question does not satisfy a given condition, a given flag is applied to the response to the question. In this manner, the response of the user is determined more specifically, whereby this processing enables, in the two-element authentication, authentication with higher accuracy.

Eighthly, in the above-described user authentication processing according to the first embodiment, the skill model of the user is updated on the basis of a received response to a question. In this manner, the skill model is refined at every user authentication processing, whereby this processing enables, in the two-element authentication, user authentication with higher accuracy.

The invention enables, in the two-element authentication, user authentication with high accuracy.

Computer Program

A computer program that describes processing executed by the user authentication device 10 according to the embodiment in a computer-executable language can be created. As one embodiment, the user authentication device 10 can be implemented by installing a user authentication computer program that executes the user authentication processing described above as packaged software and online software into a desired computer. For example, by causing an information-processing device to execute the user authentication computer program described above, the information-processing device can function as the user authentication device 10. The information-processing device mentioned here includes a desktop or a laptop personal computer. Besides, mobile communication terminals such as a smartphone, a mobile phone, and a personal handyphone system (PHS), slate terminals such as a personal digital assistant (PDA), and the like are in the category of the information-processing device. In addition, functions of the user authentication device 10 may be implemented in a cloud server.

FIG. 5 is a view illustrating an example of a computer that executes a user authentication computer program. A computer 1000 includes, for example, a memory 1010, a CPU 1020, a hard disk drive interface 1030, a disk drive interface 1040, a serial port interface 1050, a video adapter 1060, and a network interface 1070. Those units are connected one another by a bus 1080.

The memory 1010 includes a read only memory (ROM) 1011 and a RAM 1012. The ROM 1011 stores therein, for example, a boot program of a basic input output system (BIOS) and the like. The hard disk drive interface 1030 is connected to a hard disk drive 1031. The disk drive interface 1040 is connected to a disk drive 1041. For example, attachable/detachable storage media such as a magnetic disk and an optical disk are inserted into the disk drive 1041. For example, a mouse 1051 and a keyboard 1052 are connected to the serial port interface 1050. For example, a display 1061 is connected to the video adapter 1060.

The hard disk drive 1031 stores therein, for example, an operating system (OS) 1091, an application program 1092, a program module 1093, and program data 1094. Each information described in the embodiment is stored in, for example, the hard disk drive 1031 and the memory 1010.

The user authentication computer program is stored in the hard disk drive 1031 as, for example, the program module 1093 that describes commands executed by the computer 1000. Specifically, the program module 1093 that describes each processing executed by the user authentication device 10 described in the embodiment is stored in the hard disk drive 1031.

Data used in information processing by the user authentication computer program is stored as the program data 1094 in, for example, the hard disk drive 1031. The CPU 1020 reads the program module 1093 and the program data 1094 stored in the hard disk drive 1031 into the RAM 1012 as needed, and executes each procedure described above.

The program module 1093 and the program data 1094 according to the user authentication computer program are not always stored in the hard disk drive 1031, and may be stored in, for example, an attachable/detachable storage medium and be read by the CPU 1020 via the disk drive 1041 and the like. The program module 1093 and the program data 1094 according to the user authentication computer program may be stored in the other computer connected via networks such as a local area network (LAN) and a wide area network (WAN), and be read by the CPU 1020 via the network interface 1070.

Although the invention has been described with respect to specific embodiments for a complete and clear disclosure, the appended claims are not to be thus limited but are to be construed as embodying all modifications and alternative constructions that may occur to one skilled in the art that fairly fall within the basic teaching herein set forth.

Claims

1. A user authentication device comprising: a memory; anda processor coupled to the memory and programmed to execute a process comprising;collecting information of a user, wherein the information specifies a user skill;constructing, using the collected information, a skill model that has learned a category of a task to be performed by the user and the user skill, wherein the task is according to the user skill;generating, using the constructed skill model, a question for the user that causes the user to perform a task corresponding to the category and the user skill;presenting the generated question to the user via an application that outputs the question and instructing the user to perform the task, when authentication of the user is required;receiving, from the user via an application into which the response is input, an accuracy of an answer of the user to the question, a response time, or characteristics at a time of solving the question a result of performing the task as a response to the presented question;determining that the authentication of the user is successful if the accuracy of an answer to the question, the response time, and the characteristics at a time of solving the question are within a range of parameters set by the skill model;applying, by the application, a given flag indicating that the received response does not satisfy a given condition, as two-element authentication of the user, wherein an element of the two-element authentication comprises authentication of the user on the basis of the user skill of the performed task indicated by the received response, and the other element of the two-element authentication comprises authentication that is distinct from the authentication on the basis of the user skill;updating the skill model each time the user is authenticated, by inputting the accuracy of an answer of the user to the question, the response time, or the characteristics at a time of solving the question of data; andgranting intended access to data by logging into the application by indicating, by the application, on the basis of a result of the two-element authentication, a warning on a display to the user to control the intended access to data by logging into the application.

2. The user authentication device according to claim 1, further comprising collecting the information using text analysis, natural language processing, or semantic analysis.

3. The user authentication device according to claim 1, further comprising generating, as the question, a numerical question, a mathematic question, a physical question, a mechanical question, a reading exercise, or a speed chess.

4. The user authentication device according to claim 1, further comprising presenting another question if the response does not satisfy a given condition.

5. The user authentication device according to claim 1, further comprising applying a given flag to the response if the response does not satisfy a given condition.

6. A user authentication method performed by a user authentication device, the user authentication method comprising: collecting information of a user, wherein the information specifies a user skill;constructing, using the collected information, a skill model that has learned a category of a task to be performed by the user and the user skill, wherein the task is according to the user skill;generating, using the constructed skill model, a question for the user, that causes the user to perform a task corresponding to the category and the user skill;presenting the generated question to the user via an application that outputs the question and instructing the user to perform the task, when authentication of the user is required;receiving, from the user via an application into which the response is input, an accuracy of an answer of the user to the question, a response time, or characteristics at a time of solving the question as a response to the presented question;determining that the authentication of the user is successful if the accuracy of an answer to the question, the response time, and the characteristics at a time of solving the question are within a range of parameters set by the skill model;applying, by the application, a given flag indicating that the received response does not satisfy a given condition, as two-element authentication of the user, wherein an element of the two-element authentication comprises authentication of the user on the basis of the user skill of the performed task indicated by the received response, and the other element of the two-element authentication comprises authentication that is distinct from the authentication on the basis of the user skill;updating the skill model each time the user is authenticated, by inputting the accuracy of an answer of the user to the question, the response time, or the characteristics at a time of solving the question as data; andgranting intended access to data by logging into the application by indicating, by the application, on the basis of a result of the two-element authentication, a warning on a display to the user to control the intended access to data by logging into the application.

7. A non-transitory recording medium recording a user authentication computer program to be executed by a computer, the program causing the computer to perform operations comprising: collecting information of a user, wherein the information specifies a user skill;constructing, using the collected information, a skill model that has learned a category of a task to be performed by the user and the user skill, wherein the task is according to the user skill;generating, using the constructed skill model, a question for the user that causes the user to perform a task corresponding to the category and the user skill;presenting the generated question to the user via an application that outputs the question and instructing the user to perform the task, when authentication of the user is required;receiving, from the user via an application into which the response is input, an accuracy of an answer of the user to the question, a response time, or characteristics at a time of solving the question as a response to the presented question;determining that the authentication of the user is successful if the accuracy of an answer to the question, the response time, and the characteristics at a time of solving the question are within a range of parameters set by the skill model;applying, by the application, a given flag indicating that the received response does not satisfy a given condition, as two-element authentication of the user, wherein an element of the two-element authentication comprises authentication of the user on the basis of the user skill of the performed task indicated by the received response, and the other element of the two-element authentication comprises authentication that is distinct from the authentication on the basis of the user skill;updating the skill model each time the user is authenticated, by inputting the accuracy of an answer of the user to the question, the response time, or the characteristics at a time of solving the question as data; andgranting intended access to data by logging into the application by indicating, by the application, on the basis of the two-element authentication, a warning on a display to the user to control the intended access to data by logging into the application.