CN105678125B - A kind of user authen method, device - Google Patents
A kind of user authen method, device Download PDFInfo
- Publication number
- CN105678125B CN105678125B CN201410670073.1A CN201410670073A CN105678125B CN 105678125 B CN105678125 B CN 105678125B CN 201410670073 A CN201410670073 A CN 201410670073A CN 105678125 B CN105678125 B CN 105678125B
- Authority
- CN
- China
- Prior art keywords
- user
- behavioral data
- terminal
- touches
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- User Interface Of Digital Computer (AREA)
- Collating Specific Patterns (AREA)
- Telephone Function (AREA)
Abstract
The application provides a kind of user authen method, which comprises the user for receiving terminal acquisition touches behavioral data；Judge that the user touches the habit touch behavior whether behavioral data matches the authorized user of the terminal；If so, determining the user for authorized user；If not, determining the user for unauthorized user, the user that the application is acquired based on terminal touches behavioral data and carries out authentication, is not easy counterfeit, and safety is more preferable；And authentication can be completed to user in the unwitting situation of user, there is certain crypticity, therefore when terminal is lost, the leakage of personal information will not be caused.
Description
Technical field
This application involves field of terminal technology more particularly to a kind of user authen method and devices.
Background technique
In existing realization, authentication is carried out to user, is normally based on user name-password or physiological characteristic (example
Such as fingerprint) it is authenticated.
For based on user name-password identification authentication mode, due to much being applied in current mobile device, for example (,) it is micro-
Believe, know, everybody etc. social applications, usually only the identity of user is authenticated when user logs in for the first time, user only needs
An account and password are inputted, subsequent user is no longer needed for repeatedly inputting account and password when using the application, therefore works as
After mobile device is lost, these accounts and password are also at a kind of stolen status, and the personal information of user will be let out at this time
Dew, to cause damages.
For the identification authentication mode based on physiological characteristic, although highly-safe in user name-password authentication side
Then formula but exists and is easy to the shortcomings that forging, such as counterfeiter can be smooth by one fingerprint authentication equipment of purchase
By authentication, to achieve the purpose that steal userspersonal information.
Summary of the invention
To solve the problems, such as that background technique, the application provide one kind in user authen method and device.
According to the embodiment of the present application in a first aspect, the application proposes a kind of user authen method, apply in server-side, institute
The method of stating includes:
The user for receiving terminal acquisition touches behavioral data；
Judge that the user touches the habit touch behavior whether behavioral data matches the authorized user of the terminal；
If so, determining the user for authorized user；
If not, determining the user for unauthorized user.
Optionally, the user touch finger contact area, contact position, time of contact that behavioral data includes: user,
One or more combination in the information such as press pressure and glide direction.
Optionally, described to judge that the user touches the habit touching whether behavioral data matches the authorized user of the terminal
Before touching behavior, further includes:
The user for collecting the terminal acquisition touches behavioral data；
Behavioral data is touched to the user being collected into and carries out data cleansing；
The user after data cleansing is touched into behavioral data and is organized into vector, and is chosen to be white sample, and from originally
The user of the other users stored in ground database touches the black sample that preset quantity is selected in behavioral data；
Classified calculating is carried out to the selected white sample and the black sample according to default sorting algorithm, is determined described
The classifying face of white sample and the black sample.
Optionally, described to judge that the user touches the habit touching whether behavioral data matches the authorized user of the terminal
The behavior of touching includes:
The user received touch behavioral data is organized into vector；
Judge whether the vector sorted out is located at the white sample side of the classifying face；
When the vector is located at the white sample side of the classifying face, determine that the user touches described in behavioral data matching
The habit of the authorized user of terminal touches behavior；
When the vector is located at the black sample side of the classifying face, determine that the user touches behavioral data and mismatches institute
The habit for stating the authorized user of terminal touches behavior.
Optionally, the method also includes:
When the vector is located at the black sample side of the classifying face, the vector and the classifying face are further calculated
Distance；
When the distance is greater than threshold value, determine that the user touches the authorized user that behavioral data mismatches the terminal
Habit touch behavior；
When the distance is less than threshold value, the corresponding user of behavioral data is touched to the user and is further recognized
Card determines that the user touches the habit touch row that behavioral data matches the authorized user of the terminal if certification passes through
For.
Optionally, the method also includes:
According to preset period of time, the user for periodically updating the terminal acquisition being collected into touches behavioral data,
And behavioral data is touched for the updated user according to the default sorting algorithm and carries out classified calculating, to described point
Class face is updated.
According to the second aspect of the embodiment of the present application, the application proposes a kind of user authentication device, applies in server-side, institute
Stating device includes:
Receiving module touches behavioral data by the user of terminal acquisition for receiving；
Judgment module, for judging that the user touches the habit whether behavioral data matches the authorized user of the terminal
Touch behavior；
If so, determining the user for authorized user；
If not, determining the user for unauthorized user.
Optionally, the user touch finger contact area, contact position, time of contact that behavioral data includes: user,
One or more combination in the information such as press pressure and glide direction.
Optionally, described device further include:
Collection module, the user for collection terminal acquisition touch behavioral data；
Cleaning module carries out data cleansing for touching behavioral data to the user being collected into；
Categorization module is organized into vector for the user after data cleansing to be touched behavioral data, and is chosen to be white
Sample, and the black sample that preset quantity is selected in behavioral data is touched from the user of the other users stored in local data base
This；Classified calculating is carried out to the selected white sample and the black sample according to default sorting algorithm, determines the white sample
The classifying face of this and the black sample.
Optionally, the judgment module is further used for:
The user that receiving module receives is touched into behavioral data and is organized into vector, whether judges the vector sorted out
Positioned at the white sample side of the classifying face；When the vector is located at the white sample side of the classifying face, the user is determined
The habit for touching the authorized user that behavioral data matches the terminal touches behavior；When the vector is located at the black of the classifying face
Sample side determines that the user touches the habit touch behavior that behavioral data mismatches the authorized user of the terminal.
Optionally, the judgment module is further used for:
When the vector is located at the black sample side of the classifying face, the vector and the classifying face are further calculated
Distance；
When the distance is greater than threshold value, determine that the user touches the authorized user that behavioral data mismatches the terminal
Habit touch behavior；
When the distance is less than threshold value, the corresponding user of behavioral data is touched to the user and is further recognized
Card determines that the user touches the habit touch row that behavioral data matches the authorized user of the terminal if certification passes through
For.
Optionally, the collection module is further used for:
According to preset period of time, the user for periodically updating the terminal acquisition being collected into touches behavioral data；
The categorization module is further used for: touching row for the updated user according to the default sorting algorithm
Classified calculating is carried out for data, to be updated to the classifying face.
According to the third aspect of the embodiment of the present application, the application proposes a kind of user authentication device, comprising:
Processor；For storing the memory of the processor-executable instruction；
Wherein, the processor is configured to:
The user for receiving terminal acquisition touches behavioral data；
Judge that the user touches the habit touch behavior whether behavioral data matches the authorized user of the terminal；
If so, determining the user for authorized user；
If not, determining the user for unauthorized user.
The application from existing realization using based on the active certifications mode such as user name-password or physiological characteristic it is different
It is to use a kind of passive certificate scheme, server-side touches behavioral data and the end by the user for acquiring terminal on backstage
The habit of the authorized user at end touches behavior and matches, to determine whether the user is authorized user；In existing realization
User authentication mode is compared, and carries out authentication since the application touches behavioral data based on the user that terminal acquires, no
Easily counterfeit, safety is more preferable；And the application can complete authentication to user in the unwitting situation of user, have one
Fixed crypticity, therefore when terminal is lost, the leakage of personal information will not be caused.
Detailed description of the invention
Fig. 1 is a kind of flow chart for user authen method that one embodiment of the application provides；
Fig. 2 is the flow chart for another user authen method that one embodiment of the application provides；
Fig. 3 is the schematic diagram for the classifying face that one embodiment of the application provides；
Fig. 4 is the logical construction schematic diagram for the user authentication device that one embodiment of the application provides；
Fig. 5 is the structural schematic diagram of the operation carrier for the carrying user authentication device that one embodiment of the application provides.
Specific embodiment
The application proposes a kind of user authen method, touches behavioral data and the terminal by the user for acquiring terminal
The habit of authorized user touches behavior and matches, to determine whether the user is authorized user；With the user in existing realization
Authentication mode is compared, and is carried out authentication since the application touches behavioral data based on the user that terminal acquires, is not easy to imitate
It emits, safety is more preferable；And the application can complete authentication to user in the unwitting situation of user, have centainly
Crypticity, therefore when terminal is lost, the leakage of personal information will not be caused.
The application is described below by specific embodiment and in conjunction with different application example.
Referring to FIG. 1, Fig. 1 is a kind of user authen method that one embodiment of the application provides, the executing subject of this method
It can be server-side；It the described method comprises the following steps:
Step 101, the user for receiving terminal acquisition touches behavioral data；
In the present embodiment, it may include finger contact surface of the user in operating terminal that the user, which touches behavioral data,
One or more combination in the characteristic parameters such as product, contact position, time of contact, press pressure and glide direction, therefore
Terminal can be realized when acquiring above-mentioned user's touch behavioral data by the touch event of monitoring users；For example, terminal can
With the touch event of real-time monitoring users, then on backstage by the user listened in operating terminal when occur sliding, point
It hits the relevant information collection of equal touch events to get off, and is uploaded to server-side.
It is worth noting that it is logical can only to acquire those when acquiring the touch behavioral data of user for terminal when realizing
The touch behavioral data of the user tentatively authenticated is crossed, for example, terminal can carry out just user based on user name-password mode
Step certification, only acquires the touch behavioral data for the user that those have input correct username and password；Certainly, if the end
The preliminary certification or this programme is not supported (for example to be replaced hand with this programme as system-level user authen method in end
The function of machine lock), terminal can default the acquisition that all users touch with behavioral data, and be uploaded to server-side；Due to
The user uploaded at this time touches in behavioral data, may touch behavioral data comprising the user of some unauthorized users, therefore
A preliminary screening or server-side can be carried out to the data that terminal uploads in service end side to handle to data
When, the user that the user touches the unauthorized user for including in behavioral data is touched into behavioral data in the data cleansing stage and is picked
It removes.
Step 102, judge that the user touches the habit touch whether behavioral data matches the authorized user of the terminal
Behavior；If so, determining the user for authorized user；If not, determining the user for unauthorized user.
In the present embodiment, server-side touches behavioral data in the user for receiving terminal acquisition, carries out to the user
When authentication, the user of the user can be touched to the habit of behavioral data and the preset authorized user for being used to characterize the terminal
The data model of used touch behavior is matched, thus to confirm whether the user is authorized user.
The data model can be server-side for largely by user's touch behavioral data progress mould of terminal acquisition
It is obtained after type training.Server-side, can be according to pre- when touching behavioral data progress data model training for the user
If machine learning classification algorithm, for example, the machine learning classification algorithm may include SVM algorithm, GBDT algorithm, decision tree
The collected user of this terminal is touched behavioral data and is chosen to be white sample, by local data base by algorithm, KNN algorithm scheduling algorithm
The user of other users of middle storage touches behavioral data (including using the user of other users of the terminal to touch behavioral data
With the user's touch behavioral data for using other terminals) it is chosen to be black sample, when realizing, the black sample can be selected preferentially
Those touch behavioral data using the user of other users of present terminal, if the user of other users using present terminal
Touch behavioral data data volume it is less is not enough to complete to calculate, can also be stored from local data base again using other ends
The user at end touches in behavioral data and extracts a part of data as black sample.It, can after the completion of black sample and white sample are selected
To carry out classified calculating for selected white sample and black sample, so that it is determined that the classifying face of the white sample and black sample out.
Specifically, carrying out classified calculating to selected white sample and black sample according to preset machine learning classification algorithm
When, data cleansing can be carried out to selected white sample and black sample first, by some invalid and missing data or white sample
The user of some unauthorized users mixed in this touches behavioral data and rejects, and by the data after cleaning be organized into respectively to
Amount, then according to preset sorting algorithm to after being organized into vector white sample and black sample calculate, and finally determine
The classifying face of white sample and black sample.Certainly, if when realizing, the same terminal is there are when multiple authorized users, server-side
Classifying face will be calculated separately for multiple authorized user, i.e. server-side needs to distinguish for each authorized user of the same terminal
Establish a classifying face.
Wherein, the classifying face can classify to selected white sample and black sample, i.e., by server-side classification meter
White sample and black sample vector after calculation, are located at the two sides of calculated classifying face.By in this present embodiment using described
The collected user of terminal touches behavioral data as white sample, the user of the other users stored in server-side local data base
Behavioral data is touched as black sample, therefore the classifying face can be used for characterizing using the authorized user of the terminal and other use
The customary difference of the touch at family.
In the present embodiment, after server-side calculates the classifying face of the black sample and the white sample, the data
Model training finishes, and the subsequent user that can be acquired the terminal received touches behavioral data and data model progress
Match to determine using whether the user of the terminal is authorized user.
Specifically, there is still a need for what will be received to be organized by the collected user's touch behavioral data of the terminal for server-side
Then vector the vector sorted out is compared with the classifying face, if the vector is located at the white sample of the classifying face
Then should then can be determined that the user touches the habit touch behavior that behavioral data matches the authorized user of the terminal side
User is authorized user；On the contrary, can be determined that user's touching if the vector is located at the black sample side of the classifying face
The habit for touching the authorized user that behavioral data mismatches the terminal touches behavior, which is unauthorized user.Certainly, it is realizing
When, the data model can also be synchronized to terminal by server-side, confirm user since above-mentioned matching process is executed by terminal
Whether it is authorized user, is no longer described in detail.
In the present embodiment, for authorized user, can normally pass through into the system of terminal or into certification
Using progress normal operating；For unauthorized user, its system for entering terminal can be refused, or user is currently made
It removes in application in the userspersonal information that system background saves, and user is authenticated again；For example, this programme can
With the user authentication for system level, for example it may be substituted for the function of cell-phone lock, when user is judged as unauthorized use
Behind family, mobile phone can be directly locked；This programme can also carry out user authentication for some specific applications installed in terminal,
For example Alipay wallet can be close by the user name saved in Alipay wallet after user is judged as unauthorized user
Code is removed, and user is prompted to re-enter account number cipher.
In the embodiment above, server-side touches behavioral data by the user that collection terminal uploads, according to what is be collected into
The user that user touches the other users stored in behavioral data and local data base touches behavioral data and carries out classification based training,
The data model of the habit touch behavior for the authorized user that can characterize terminal is trained, and when authenticating to user, it will
Terminal acquisition user touch behavioral data matched with the data model, by judge the user operation behavior whether
The habit for matching authorized user touches behavior, to determine whether the user is authorized user；With the user authentication in existing realization
Mode is compared, and is carried out authentication since the application touches behavioral data based on the user that terminal acquires, is not easy counterfeit, safety
More preferably；And the application can complete authentication to user in the unwitting situation of user, have certain crypticity, because
This will not cause the leakage of personal information when terminal is lost.
Referring to FIG. 2, Fig. 2 is a kind of user authen method that one embodiment of the application provides, the executing subject of this method
It can be server, the server can be server, server cluster and the cloud platform that authentication is carried out to user；
It the described method comprises the following steps:
The user that step 201, collection terminal acquire touches behavioral data；
Step 202 touches behavioral data progress data cleansing to the user being collected into；
User touch behavioral data after data cleansing is organized into vector, and is chosen to be white sample by step 203,
And the black sample that preset quantity is selected in behavioral data is touched from the user of the other users stored in local data base；
Step 204 carries out classified calculating to the selected white sample and the black sample according to default sorting algorithm, really
Make the classifying face of the white sample and the black sample；
Step 205, the user for receiving terminal acquisition touch behavioral data, and the user received is touched behavioral data and is arranged
At vector, and judge whether the vector is located at the white sample side of the classifying face；
Step 206, when the vector is located at the white sample side of the classifying face, determine the user for authorized user；When
The vector is located at the black sample side of the classifying face, determines the user for unauthorized user.
In the present embodiment, it may include finger contact surface of the user in operating terminal that the user, which touches behavioral data,
One or more combination in the information such as product, contact position, time of contact, press pressure and glide direction, therefore terminal
When acquiring user's touch behavioral data, can be realized by the touch event of monitoring users；For example, terminal can be real
When monitoring users touch event, then on backstage by the user listened in operating terminal when sliding, the click etc. that occur
The relevant information collection of touch event is got off, and is uploaded to server-side.
It is worth noting that terminal can be based on user name-password when acquiring the behavioral data of user when realizing
Mode user is tentatively authenticated, only acquire the touch behavior number for the user that those have input correct username and password
According to；For example, it is assumed that this programme is used to carry out user identity authentication to some specific applications installed in terminal, for example this programme can
To be used to carry out user identity authentication to similar to payment application as Alipay wallet, therefore terminal is in the touching for acquiring user
When touching behavioral data, the behavioral data for the user that those have input correct username and password can be only acquired, then will be adopted
The data collected are uploaded to server.When realizing, this programme is also used as system-level user authen method, for example utilizes
The application replaces the function of cell-phone lock, therefore for this system-level application scenarios, terminal can be defaulted to all uses
The user of the terminal touch the acquisition of behavioral data, and is uploaded to server；Since under the application scenarios, terminal is uploaded
User touch in behavioral data, behavioral data may be touched comprising the user of some unauthorized users, therefore can taken
Business device side carries out a preliminary screening to the data that terminal uploads, or when server handles data, in data
The user that the user touches the unauthorized user for including in behavioral data is touched behavioral data and rejected by wash phase.
In the present embodiment, server touches behavioral data in the user for receiving terminal acquisition, carries out to the user
When authentication, the user of the user can be touched to the habit of behavioral data and the preset authorized user for being used to characterize the terminal
The data model of used touch behavior is matched, to confirm whether the user is authorized user.
Wherein, the data model can be server for largely by user's touch behavioral data of terminal acquisition
It is obtained after progress model training.Server, can be with when touching behavioral data for the user and carrying out data model training
According to preset machine learning classification algorithm, the collected user of terminal is touched into behavioral data and is chosen to be white sample, it will be local
The user of the other users stored in database touches behavioral data (including using the user of other users of the terminal to touch row
Behavioral data is touched for data and using the user of other terminals) it is chosen to be black sample, when realizing, the black sample can be excellent
It first selectes those and touches behavioral data using the user of other users of present terminal, if using other users of present terminal
User touch behavioral data data volume it is less be not enough to complete to calculate, the use that can also be stored from local data base again
The user of other terminals touches in behavioral data and extracts a part of data as black sample.When black sample and the selected completion of white sample
Afterwards, classified calculating can be carried out for selected white sample and black sample, so that it is determined that point of the white sample and black sample out
Class face.Wherein, Fig. 3 is referred to, the classifying face can classify to selected white sample and black sample, that is, pass through server
White sample and black sample vector after classified calculating, are located at the two sides of calculated classifying face.The machine learning classification
Algorithm, main technology and algorithm comprising classifying to data, such as the machine learning classification algorithm may include that SVM is calculated
Method, GBDT algorithm, decision Tree algorithms, KNN algorithm scheduling algorithm effectively can make modeling to data and carry out data classification, and have very well
Prediction effect；By touching behavioral data as white sample, service using the collected user of the terminal in this present embodiment
The user of the other users stored in device local data base touches behavioral data as black sample, therefore the classifying face can be used
The customary difference of touch of the authorized user and other users of the terminal are used in characterization.
In the present embodiment, server according to preset machine learning classification algorithm to selected white sample and black sample
When carrying out classified calculating, data cleansing can be carried out to selected white sample and black sample first, it will be some invalid and missing
The users of some unauthorized users mixed in data or white sample touches behavioral data and rejects, and by the data after cleaning
It is organized into vector respectively, is then calculated according to preset sorting algorithm based on the vector sorted out, and finally determine white
The classifying face of sample and black sample.Certainly, if when realizing, for the same terminal there are when multiple authorized users, server-side will
Classifying face is calculated separately for multiple authorized user, i.e. server-side needs to build respectively for each authorized user of the same terminal
Found a classifying face.
Wherein, it is worth noting that, server arranges white sample and black sample after becoming vector, and the vector sorted out can
To be multi-C vector；For example, as can be seen from the above description, the user, which touches behavioral data, in the present embodiment can be user
In the characteristic parameters such as finger contact area, contact position, time of contact, press pressure and glide direction in operating terminal
One or more combination, therefore when the user touches the combination that behavioral data is multiple characteristic parameters, it arranges at this time
Vector out will be rendered into a kind of form of multi-C vector, and theoretically the dimension of the vector is higher, and server is by classification meter
The data model trained after calculation is then more accurate, therefore when realizing, the user touches behavioral data can be as far as possible
The combination for selecting different characteristic parameters.
In the present embodiment, after server calculates the classifying face of the black sample and the white sample, the data
Model training finishes, and the subsequent user that can be acquired the terminal received touches behavioral data and data model progress
Match to determine using whether the user of the terminal is authorized user.
Specifically, there is still a need for what will be received to be organized by the collected user's touch behavioral data of the terminal for server
Then vector the vector sorted out is compared with the classifying face, if the vector is located at the white sample of the classifying face
Then should then can be determined that the user touches the habit touch behavior that behavioral data matches the authorized user of the terminal side
User is authorized user；On the contrary, can be determined that user's touching if the vector is located at the black sample side of the classifying face
The habit for touching the authorized user that behavioral data mismatches the terminal touches behavior, which is unauthorized user.Certainly, it is realizing
When, the data model can also be synchronized to terminal by server, confirm user since above-mentioned matching process is executed by terminal
Whether it is authorized user, is no longer described in detail.
In the present embodiment, for authorized user, can normally pass through into the system of terminal or into certification
Using progress normal operating；For unauthorized user, its system for entering terminal can be refused, or user is currently made
It removes in application in the userspersonal information that system background saves, and user is authenticated again；For example, this programme can
With the user authentication for system level, for example it may be substituted for the function of cell-phone lock, when user is judged as unauthorized use
Behind family, mobile phone can be directly locked；This programme can also carry out user authentication for some specific applications installed in terminal,
For example Alipay wallet can be close by the user name saved in Alipay wallet after user is judged as unauthorized user
Code is removed, and user is prompted to re-enter account number cipher.
In the embodiment above, server-side touches behavioral data by the user that collection terminal uploads, according to what is be collected into
The user that user touches the other users stored in behavioral data and local data base touches behavioral data and carries out classification based training,
The data model of the habit touch behavior for the authorized user that can characterize terminal is trained, and when authenticating to user, it will
Terminal acquisition user touch behavioral data matched with the data model, by judge the user operation behavior whether
The habit for matching authorized user touches behavior, to determine whether the user is authorized user；With the user authentication in existing realization
Mode is compared, and is carried out authentication since the application touches behavioral data based on the user that terminal acquires, is not easy counterfeit, safety
More preferably；And the application can complete authentication to user in the unwitting situation of user, have certain crypticity, because
This will not cause the leakage of personal information when terminal is lost.
As previously mentioned, in practical applications, user's touch behavioral data that terminal uploads is likely to mix some unauthorized
The touch behavioral data of user, and, in some special scenes, such as crowded public arena, user is in operating terminal
When, it is possible that some be different from the abnormal operation for touching habit, behavior number is touched so as to cause the user that terminal is collected into
According to the operating habit for deviateing user, therefore for this kind of data, if server-side fails to reject in time during data cleansing,
It may result in server to judge by accident when carrying out authentication to user.Therefore, in another alternative embodiment, exist
On the basis of above embodiments, the user received touch behavioral data is being organized into vector by server, and will be sorted out
It, can be further if the vector is located at the black sample side of the classifying face when classifying face described in vector sum is compared
The vector is calculated at a distance from the classifying face, if calculated distance is greater than threshold value, can directly determine that the user is
Unauthorized user；If calculated distance is less than threshold value, the vector and the classifying face are very close at this time, it is likely that deposit
It is judging by accident, therefore the user can further authenticated, for example user name-password or other certifications can be based on
Mode authenticates the user, if certification passes through, it is possible to determine that the user is authorized user.
In the embodiment above, it is located at the black of the classifying face by touching the corresponding vector of behavioral data in the user
When sample side, the vector is calculated at a distance from classifying face, and when calculated distance is less than threshold value, which is carried out into one
The certification of step, and after certification passes through, determine that the user for authorized user, can effectively reduce the erroneous judgement of server-side.
In another alternative embodiment, on the basis of above embodiments, since user is in the process of using terminal
In, over time, the operating habit of user can also may change over time, therefore server-side may be used also
To set a modeling period (preset period of time), periodically the data model is carried out more according to modeling period
Newly；Specifically, server-side can periodically update the user uploaded by terminal being collected into and touch behavior according to modeling period
Data, and behavioral data progress classified calculating is touched for the updated user according to default sorting algorithm and obtains classification
Face is updated with calculating the classifying face to a upper modeling period, so as to avoid the operating habit of user with
After the passage of time changes, server-side is caused to judge by accident.
In the embodiment above, by setting modeling period, and modeling period periodically carries out the data model
Update, can to avoid the operating habit as user changes over time and caused by server-side judge by accident.
Corresponding with above method embodiment, present invention also provides a kind of embodiments of user authentication device.It refers to
Fig. 4, the application also propose a kind of user authentication device 40, are applied to server-side, such as the server-side can be server；Its
In, refer to Fig. 5, as the operation carrier of the application device 40, the server-side typically at least include CPU, memory, it is non-easily
The property lost memory, network interface and internal bus etc.；Taking software implementation as an example, the application device 40 is it is generally understood that add
Computer program in memory is carried, the logic device combined by the software and hardware formed after CPU operation；Described device 40
Include:
Receiving module 401 touches behavioral data by the user of terminal acquisition for receiving；
Judgment module 402, for judging whether user's touch behavioral data matches the authorized user's of the terminal
It is accustomed to touch behavior；If so, determining the user for authorized user；If not, determining the user for unauthorized user.
In the present embodiment, described device further include:
Collection module 403, the user for collection terminal acquisition touch behavioral data；
Cleaning module 404 carries out data cleansing for touching behavioral data to the user being collected into；
Categorization module 405 is organized into vector for the user after data cleansing to be touched behavioral data, and is chosen to be
White sample, and the black sample that preset quantity is selected in behavioral data is touched from the user of the other users stored in local data base
This；Classified calculating is carried out to the selected white sample and the black sample according to default sorting algorithm, determines the white sample
The classifying face of this and the black sample.
In the present embodiment, the judgment module 402 is further used for:
The user that receiving module receives is touched into behavioral data and is organized into vector, whether judges the vector sorted out
Positioned at the white sample side of the classifying face；When the vector is located at the white sample side of the classifying face, the user is determined
The habit for touching the authorized user that behavioral data matches the terminal touches behavior；When the vector is located at the black of the classifying face
Sample side determines that the user touches the habit touch behavior that behavioral data mismatches the authorized user of the terminal.
In the present embodiment, the judgment module 402 is further used for:
When the vector is located at the black sample side of the classifying face, the vector and the classifying face are further calculated
Distance；
When the distance is greater than threshold value, determine that the user touches the authorized user that behavioral data mismatches the terminal
Habit touch behavior；
When the distance is less than threshold value, the corresponding user of behavioral data is touched to the user and is further recognized
Card determines that the user touches the habit touch row that behavioral data matches the authorized user of the terminal if certification passes through
For.
In the present embodiment, the collection module 403 is further used for:
According to preset period of time, the user for periodically updating the terminal acquisition being collected into touches behavioral data；
The categorization module is further used for: touching row for the updated user according to the default sorting algorithm
Classified calculating is carried out for data, to be updated to the classifying face.
In the present embodiment, the user touches behavioral data and includes: the finger contact area of user, contact position, connects
Touch one or more the combination in the information such as time, press pressure and glide direction.
For device embodiment, since it corresponds essentially to embodiment of the method, so related place is referring to method reality
Apply the part explanation of example.The apparatus embodiments described above are merely exemplary, wherein described be used as separation unit
The module of explanation may or may not be physically separated, and the component shown as module can be or can also be with
It is not physical module, it can it is in one place, or may be distributed on multiple network modules.It can be according to actual
The purpose for needing to select some or all of the modules therein to realize application scheme.Those of ordinary skill in the art are not paying
Out in the case where creative work, it can understand and implement.
The foregoing is merely the preferred embodiments of the application, not to limit the application, all essences in the application
Within mind and principle, any modification, equivalent substitution, improvement and etc. done be should be included within the scope of the application protection.
Present invention also provides a kind of embodiments of server.
The server includes: processor, and the memory for storing the processor-executable instruction；Further,
The server can also include input/output interface, network interface, various hardware etc..
The server can be configured as:
The user for receiving terminal acquisition touches behavioral data；
Judge that the user touches the habit touch behavior whether behavioral data matches the authorized user of the terminal；
If so, determining the user for authorized user；
If not, determining the user for unauthorized user.
Those skilled in the art after considering the specification and implementing the invention disclosed here, will readily occur to its of the application
Its embodiment.This application is intended to cover any variations, uses, or adaptations of the application, these modifications, purposes or
Person's adaptive change follows the general principle of the application and including the undocumented common knowledge in the art of the application
Or conventional techniques.The description and examples are only to be considered as illustrative, and the true scope and spirit of the application are by following
Claim is pointed out.
It should be understood that the application is not limited to the precise structure that has been described above and shown in the drawings, and
And various modifications and changes may be made without departing from the scope thereof.Scope of the present application is only limited by the accompanying claims.
Claims (9)
1. a kind of user authen method is applied in server-side, which is characterized in that the described method includes:
The user for receiving terminal acquisition touches behavioral data；
The user is touched the corresponding vector of behavioral data to be compared with classifying face；Wherein, the classifying face be will be described
The corresponding vector of the collected user behavior data of terminal is as white sample, by the use of the other users stored in local data base
Family touches the corresponding vector of behavioral data as black sample and carries out classified calculating, the white sample and the black sample determined
Classifying face；
If the user touches the white sample side that the corresponding vector of behavioral data is located at the classifying face, determine that the user is
The authorized user of the terminal；
If the user touches the black sample side that the corresponding vector of behavioral data is located at the classifying face, institute is further calculated
Vector is stated at a distance from the classifying face；When the distance is greater than threshold value, determine that the user is the unauthorized use of the terminal
Family；When the distance is less than threshold value, the corresponding user of behavioral data is touched to the user and is further authenticated, if
Certification passes through, then determines that the user is the authorized user of the terminal.
2. the method according to claim 1, wherein the user touches the finger that behavioral data includes: user
One or more combination in the information such as contact area, contact position, time of contact, press pressure and glide direction.
3. the method according to claim 1, wherein the method also includes:
The user for collecting the terminal acquisition touches behavioral data；
Behavioral data is touched to the user being collected into and carries out data cleansing；
The user after data cleansing is touched into behavioral data and is organized into vector, and is chosen to be white sample, and from local number
The black sample that preset quantity is selected in behavioral data is touched according to the user of the other users stored in library；
Classified calculating is carried out to the selected white sample and the black sample according to default sorting algorithm, determines the white sample
The classifying face of this and the black sample.
4. according to the method described in claim 3, it is characterized in that, the method also includes:
According to preset period of time, the user for periodically updating the terminal acquisition being collected into touches behavioral data, and root
Behavioral data is touched for the updated user according to the default sorting algorithm and carries out classified calculating, to the classifying face
It is updated.
5. a kind of user authentication device is applied in server-side, which is characterized in that described device includes:
Receiving module touches behavioral data by the user of terminal acquisition for receiving；
Judgment module is compared for the user to be touched the corresponding vector of behavioral data with classifying face；Wherein, described point
Class face is using the corresponding vector of the collected user behavior data of the terminal as white sample, by what is stored in local data base
If the user of other users touches the corresponding user of behavioral data and touches the corresponding vector of behavioral data positioned at described point
The white sample side in class face determines that the user is the authorized user of the terminal；If it is corresponding that the user touches behavioral data
Vector be located at the black sample side of the classifying face, further calculate the vector at a distance from the classifying face；When described
When distance is greater than threshold value, determine that the user is the unauthorized user of the terminal；When the distance is less than threshold value, to the use
Family touches the corresponding user of behavioral data and is further authenticated, if certification passes through, determines that the user is the end
The authorized user at end.
6. device according to claim 5, which is characterized in that the user touches the finger that behavioral data includes: user
One or more combination in the information such as contact area, contact position, time of contact, press pressure and glide direction.
7. device according to claim 5, which is characterized in that described device further include:
Collection module, the user for collection terminal acquisition touch behavioral data；
Cleaning module carries out data cleansing for touching behavioral data to the user being collected into；
Categorization module is organized into vector for the user after data cleansing to be touched behavioral data, and is chosen to be white sample,
And the black sample that preset quantity is selected in behavioral data is touched from the user of the other users stored in local data base；According to
Default sorting algorithm carries out classified calculating to the selected white sample and the black sample, determines the white sample and described
The classifying face of black sample.
8. device according to claim 7, which is characterized in that the collection module is further used for:
According to preset period of time, the user for periodically updating the terminal acquisition being collected into touches behavioral data；
The categorization module is further used for: touching behavior number for the updated user according to the default sorting algorithm
According to classified calculating is carried out, to be updated to the classifying face.
9. a kind of user authentication device characterized by comprising
Processor；For storing the memory of the processor-executable instruction；
Wherein, the processor is configured to:
The user for receiving terminal acquisition touches behavioral data；
The user is touched the corresponding vector of behavioral data to be compared with classifying face；Wherein, the classifying face be will be described
The corresponding vector of the collected user behavior data of terminal is as white sample, by the use of the other users stored in local data base
Family touches the corresponding vector of behavioral data as black sample and carries out classified calculating, the white sample and the black sample determined
Classifying face；If the user touches the white sample side that the corresponding vector of behavioral data is located at the classifying face, determining should
User is the authorized user of the terminal；If the user touches behavioral data, corresponding vector is located at the black of the classifying face
Sample side further calculates the vector at a distance from the classifying face；When the distance is greater than threshold value, the user is determined
For the unauthorized user of the terminal；When the distance is less than threshold value, the corresponding user of behavioral data is touched to the user
It is further authenticated, if certification passes through, determines that the user is the authorized user of the terminal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410670073.1A CN105678125B (en) | 2014-11-20 | 2014-11-20 | A kind of user authen method, device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410670073.1A CN105678125B (en) | 2014-11-20 | 2014-11-20 | A kind of user authen method, device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105678125A CN105678125A (en) | 2016-06-15 |
| CN105678125B true CN105678125B (en) | 2019-02-19 |
Family
ID=56958016
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410670073.1A Active CN105678125B (en) | 2014-11-20 | 2014-11-20 | A kind of user authen method, device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105678125B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US12039021B2 (en) | 2019-03-07 | 2024-07-16 | British Telecommunications Public Limited Company | Multi-level classifier based access control |
Families Citing this family (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107786528B (en) * | 2016-08-31 | 2020-09-22 | 阿里巴巴集团控股有限公司 | Application login method and device and communication system |
| CN106791195A (en) * | 2017-02-20 | 2017-05-31 | 努比亚技术有限公司 | A kind of operation processing method and device |
| CN107194216A (en) * | 2017-05-05 | 2017-09-22 | 中南大学 | A kind of mobile identity identifying method and system of the custom that swiped based on user |
| CN107944871A (en) * | 2017-11-16 | 2018-04-20 | 中国银行股份有限公司 | Identity identifying method, device, computer equipment and computer-readable recording medium |
| CN109800548B (en) * | 2017-11-17 | 2021-02-02 | 深圳市鹰硕技术有限公司 | Method and device for preventing personal information from being leaked |
| WO2020024117A1 (en) * | 2018-07-31 | 2020-02-06 | 华为技术有限公司 | Fingerprint authentication method and device, storage medium, and computer program product |
| CN109635622A (en) * | 2018-09-28 | 2019-04-16 | 阿里巴巴集团控股有限公司 | Personal identification method, device and electronic equipment |
| CN109462691B (en) * | 2018-10-27 | 2021-01-26 | 中国人民解放军战略支援部队信息工程大学 | Implicit protection method and system based on multi-sensor data fusion |
| EP3935525A1 (en) * | 2019-03-07 | 2022-01-12 | British Telecommunications public limited company | Access control |
| US10885160B1 (en) | 2019-08-21 | 2021-01-05 | Advanced New Technologies Co., Ltd. | User classification |
| CN110516418A (en) * | 2019-08-21 | 2019-11-29 | 阿里巴巴集团控股有限公司 | A kind of operation user identification method, device and equipment |
| CN110795708A (en) * | 2019-10-25 | 2020-02-14 | 支付宝(杭州)信息技术有限公司 | Security authentication method and related device |
| CN110730459B (en) * | 2019-10-25 | 2021-05-28 | 支付宝(杭州)信息技术有限公司 | Method and related device for initiating near field communication authentication |
| CN110795722A (en) * | 2019-10-25 | 2020-02-14 | 支付宝(杭州)信息技术有限公司 | Deployment method and device of security authentication model and electronic equipment |
| CN110909327A (en) * | 2019-11-08 | 2020-03-24 | 支付宝(杭州)信息技术有限公司 | Abnormity detection model updating method and device and electronic equipment |
| CN111199027B (en) * | 2019-12-30 | 2023-02-28 | 航天信息股份有限公司 | User authentication method and device, computer-readable storage medium and electronic equipment |
| CN111935162A (en) * | 2020-08-14 | 2020-11-13 | 山东云海国创云计算装备产业创新中心有限公司 | Cloud file access method and device and related components |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102203794A (en) * | 2011-06-01 | 2011-09-28 | 华为终端有限公司 | A terminal authentication method and a device thereof |
| CN102637105A (en) * | 2011-02-15 | 2012-08-15 | 鸿富锦精密工业（深圳）有限公司 | Screen unlocking method |
| CN103678965A (en) * | 2012-09-14 | 2014-03-26 | 百度在线网络技术（北京）有限公司 | Method and device for protecting safety of mobile equipment |
| CN104007929A (en) * | 2014-05-26 | 2014-08-27 | 南京泰锐斯通信科技有限公司 | Gesture recognition based mobile terminal unlocking method and mobile terminal |
| CN104200147A (en) * | 2014-09-04 | 2014-12-10 | 王洋 | Identity authentication method and system based on touch screen equipment and security and privacy encryption method |
| CN104318138A (en) * | 2014-09-30 | 2015-01-28 | 杭州同盾科技有限公司 | Method and device for verifying identity of user |
-
2014
- 2014-11-20 CN CN201410670073.1A patent/CN105678125B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102637105A (en) * | 2011-02-15 | 2012-08-15 | 鸿富锦精密工业（深圳）有限公司 | Screen unlocking method |
| CN102203794A (en) * | 2011-06-01 | 2011-09-28 | 华为终端有限公司 | A terminal authentication method and a device thereof |
| CN103678965A (en) * | 2012-09-14 | 2014-03-26 | 百度在线网络技术（北京）有限公司 | Method and device for protecting safety of mobile equipment |
| CN104007929A (en) * | 2014-05-26 | 2014-08-27 | 南京泰锐斯通信科技有限公司 | Gesture recognition based mobile terminal unlocking method and mobile terminal |
| CN104200147A (en) * | 2014-09-04 | 2014-12-10 | 王洋 | Identity authentication method and system based on touch screen equipment and security and privacy encryption method |
| CN104318138A (en) * | 2014-09-30 | 2015-01-28 | 杭州同盾科技有限公司 | Method and device for verifying identity of user |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US12039021B2 (en) | 2019-03-07 | 2024-07-16 | British Telecommunications Public Limited Company | Multi-level classifier based access control |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105678125A (en) | 2016-06-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105678125B (en) | A kind of user authen method, device | |
| KR102038851B1 (en) | Method and system for verifying identities | |
| JP5160911B2 (en) | User authentication device, user authentication method, and user authentication program | |
| US9509688B1 (en) | Providing malicious identity profiles from failed authentication attempts involving biometrics | |
| CN106650350B (en) | Identity authentication method and system | |
| CN106384027A (en) | User identity recognition system and recognition method thereof | |
| US10366217B2 (en) | Continuous user authentication | |
| KR102167602B1 (en) | Trusted terminal verification method and apparatus | |
| CN104426884A (en) | Method for authenticating identity and device for authenticating identity | |
| CN109600336A (en) | Store equipment, identifying code application method and device | |
| CN103745148A (en) | Information protection method based on fingerprint recognition and mobile terminal | |
| CN104392160A (en) | Identity authentication method and identity authentication device | |
| CN109558918A (en) | A kind of certificate chain real estate electronics license system and its application method | |
| CN110120928A (en) | A kind of identity authentication method, device, server and computer-readable medium | |
| CN105868610A (en) | Method and system for realizing user authentication through biological characteristic information | |
| CN110955677A (en) | Identity verification method, device and system | |
| CN104486306B (en) | Identity authentication method is carried out based on finger hand vein recognition and cloud service | |
| US10965696B1 (en) | Evaluation of anomaly detection algorithms using impersonation data derived from user data | |
| WO2022245581A1 (en) | Methods and systems for facilitating secure authentication of users based on known data | |
| CN105450412B (en) | Identity identifying method and device | |
| WO2016062200A1 (en) | Fingerprint authentication method and apparatus, and server | |
| US10511585B1 (en) | Smoothing of discretized values using a transition matrix | |
| CN107615299B (en) | For assessing the method and system of fingerprint template | |
| Yang et al. | Retraining and dynamic privilege for implicit authentication systems | |
| US10255558B1 (en) | Managing knowledge-based authentication systems |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: DE Ref document number: 1225823 Country of ref document: HK |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20200922 Address after: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman Islands Patentee after: Innovative advanced technology Co.,Ltd. Address before: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman Islands Patentee before: Advanced innovation technology Co.,Ltd. Effective date of registration: 20200922 Address after: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman Islands Patentee after: Advanced innovation technology Co.,Ltd. Address before: A four-storey 847 mailbox in Grand Cayman Capital Building, British Cayman Islands Patentee before: Alibaba Group Holding Ltd. |
|
| TR01 | Transfer of patent right |