CN116028908A - Continuous identity authentication method and related device based on incremental learning and meta learning - Google Patents
Continuous identity authentication method and related device based on incremental learning and meta learning Download PDFInfo
- Publication number
- CN116028908A CN116028908A CN202310163032.2A CN202310163032A CN116028908A CN 116028908 A CN116028908 A CN 116028908A CN 202310163032 A CN202310163032 A CN 202310163032A CN 116028908 A CN116028908 A CN 116028908A
- Authority
- CN
- China
- Prior art keywords
- data
- learning
- model
- authentication
- sensor
- 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.)
- Pending
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Collating Specific Patterns (AREA)
Abstract
A continuous identity authentication method and a related device based on incremental learning and meta learning comprise the following steps: performing visual operation on the sensor data, and judging whether the sensor data has a drift phenomenon or not; constructing an authentication framework comprising an offline registration stage and an online authentication stage and based on meta learning and incremental learning; acquiring the characteristics of the time dimension of each sensor and the associated characteristics among different dimensions of each sensor according to offline registration, learning from the acquired data according to an online authentication stage, and updating an online model; and performing experimental verification on the updated model to obtain the model capacity. The technical framework MetaAuth of identity authentication based on the touch screen behavior solves the problem in the field of long-term touch screen identity authentication, and secondly designs an incremental learning problem under the scene that an online updating mechanism AMUM based on meta learning is used for continuously carrying out identity authentication for a long time so as to improve the stability of a model.
Description
Technical Field
The invention relates to the technical field of deep learning, in particular to a continuous identity authentication method and a related device based on incremental learning and meta learning.
Background
In the current society, mobile phones and tablet computers are becoming the most basic tools for social, entertainment, communication and electronic commerce, and are an integral part of the current society. Currently, 83.72% of people all over the world already own smartphones, and with the continuous development of technology, the computing power and storage power of smartphones are also being continuously developed, so that more and more apps are developed and put into use, for example, in relation to social, online shopping, financial lending types, etc. According to statistics, about 60% of internet users between 16 and 64 years old make online shopping, the current annual online purchase amount is about 3.8 trillion, the average rise is 18%, and more payment behaviors can be seen to be performed at the mobile phone end.
The high reliance of people on smartphones has also led to a wide range of concerns about everyone's private data in their smartphones. Personal privacy data includes bank accounts, health, employment, etc. in the cell phone, as well as confidential data related to some government, which are recorded in the cell phone. In the general case, these data are protected and only available to the user himself. However, the data show that about one third of the mobile phones of the users are stolen, and one tenth of the users consider that the privacy in the mobile phones is infringed to different degrees. Therefore, the problem of mobile phone privacy leakage caused by various reasons becomes a great difficulty which puzzles all users and manufacturers of large mobile phones.
In order to solve the problem of mobile phone privacy, a great number of identity authentication problems have been performed to protect mobile phone private data. The mobile phone identity authentication problem mainly refers to that the mobile phone terminal verifies the identity of the user through a shorter identity indicator. So that the user can either continue access or deny access. With the research on the problems for many years, there are mainly three authentication modes: an authentication mode based on knowledge, an identity authentication mode based on biological characteristics and an identity authentication mode based on behavior characteristics; knowledge-based authentication is a relatively traditional way and is widely accepted, wherein a series of ways such as digital passwords, graphic unlocking and the like are representative. Because these modes are simpler, the cost requirement for learning is lower, so the method is greatly popularized. However, some researches show that if the password is relatively complex, the memory cost of the user is greatly increased, while if some relatively simple passwords such as abc or 12345678 are selected, the security performance is insufficient, and the user often uses the mobile phone, and he recognizes that the user uses stains in the screen or grease of the skin to perform decoding, so that the traditional mode has obvious hidden danger.
Disclosure of Invention
The invention aims to provide a continuous identity authentication method and a related device based on incremental learning and meta learning, which are used for solving the problems of safety and high memory cost in the traditional mode.
In order to achieve the above purpose, the present invention adopts the following technical scheme:
the continuous identity authentication method based on incremental learning and meta learning comprises the following steps:
collecting sensor data of the mobile phone in the using process, performing visual operation on the sensor data, and judging whether the sensor data has a drifting phenomenon or not;
when drift phenomenon occurs, an authentication framework MetaAuth comprising an offline registration stage and an online authentication stage and based on meta learning and incremental learning is constructed;
establishing a meta learning model, acquiring the characteristics of the time dimension of each sensor and the association characteristics among different dimensions of each sensor according to offline registration, learning from the acquired data according to an online authentication stage, and updating the online model;
and performing experimental verification on the updated model to obtain the model capacity.
Further, sensor data of the mobile phone in the using process are collected, visual operation is carried out on the sensor data, and whether the sensor data have drifting phenomenon or not is judged:
the method comprises the steps of randomly extracting sensor data when a mobile phone of a user is used in a certain period, dividing the period into a front period, a middle period and a later period, then extracting features of the data, respectively visualizing the features in a two-dimensional space through a t-SNE method after the features are extracted, showing the conditions of similar and changing features among the users, and when the feature data among the users change and the distance is further and further away, drifting occurs.
Further, the offline registration stage:
and (3) data acquisition: data acquisition is carried out by selecting the touch screen behaviors of the user, when the screen is lightened, the data acquired by the sensor is taken as a positive sample, and the data of other users are taken as a negative sample for training;
data preprocessing: firstly, dividing data into a series of time periods with repetition by using a fixed-length sliding window, taking the time periods as a basic unit of authentication, then continuously dividing each time period into disjoint small time periods, namely time sources, and extracting local features from the time sources;
continuous collector part: and establishing a two-channel deep learning model 2-GCTN model, and extracting the characteristics of each sensor time dimension and the associated characteristics among different dimensions.
Further, the online authentication stage:
the on-line model update is carried out at this stage by using an improved meta learning method AMUM, which comprises two parts respectively consisting of: 1. a drift detector; 2. improved MAML update policies; in the detection process, the detected time window can be changed according to the time and the change of the data, a drift signal is output according to a prediction result in the window, and if the probability of detecting the error line exceeds a set threshold value, the judgment of drift is given.
Further, the specific process comprises the following steps:
firstly, acquiring experimental data comprising three-dimensional data of a sensor from an accelerometer, a gyroscope and a magnetometer;
setting an evaluation index: the accuracy in the scene of identity authentication is measured through indexes, wherein the indexes comprise:
equal error rate: the false acceptance rate is equal to the value of the false rejection rate, and the equal false rate can be affected by changing the threshold of the prediction score;
error acceptance rate: defining the number of illegal samples received by error and the number of all illegal test samples; a high false acceptance rate indicates a high false separation rate of the intruder sample;
false rejection rate: defining the number of valid samples and the number of all valid test samples which are refused by mistake; a high false reject rate indicates poor identification of legitimate user samples;
authentication accuracy variance: a variance measure representing the degree of dispersion of authentication results over a long period of time; it is calculated from the average of the squares of the differences between the equal error rate and the average of the equal error rates for each authentication; the low variance reflects the stable predictive ability of the authentication model;
and (3) using a memory: representing the memory size occupied by model training and updating;
time consumption: representing the time it takes for the model to train and update.
Further, the test verifies that:
evaluating an authentication framework 2-GCTN through M users in a data set, independently training a model and performing incremental test on each user, setting a mobile phone corresponding to each user as a legal user, and setting the rest M-1 as illegal users; selecting data of a sensor from each user according to the sequence of time for the data of the positive sample, and dividing the data into 20 copies, wherein the 20 copies correspond to the incremental learning tasks respectively; negative samples, randomly extracted among the remaining M-1 users; in the training process, each training and test is defined as a task, and each time the data block of the next period is firstly used for testing the model, and then the training and updating of the model are carried out.
Further, the method comprises the following specific steps:
ordering all processed data according to time and dividing the processed data into identical data blocks T;
each data block will be a positive sample data set of the user, there are T-1 tasks, each task containing data D for two periods i D (D) i+1 ；
Using D i Updating the model as a training set, then using D i+1 Testing its ability;
the above steps are repeated for each user.
Further, a continuous identity authentication system based on incremental learning and meta learning includes:
the data processing judging module is used for collecting sensor data of the mobile phone in the using process, carrying out visual operation on the sensor data and judging whether the sensor data has a drifting phenomenon or not;
the framework building module is used for building an authentication framework MetaAuth comprising an offline registration stage and an online authentication stage and based on meta learning and incremental learning when a drift phenomenon occurs;
the model building module is used for building a meta learning model, acquiring the characteristics of the time dimension of each sensor and the association characteristics among different dimensions of the sensor according to offline registration, learning from the acquired data according to an online authentication stage, and updating the online model;
and the verification module is used for carrying out experimental verification on the updated model to obtain the model capacity.
Further, a computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the steps of a continuous identity authentication method based on incremental learning and meta learning when the computer program is executed.
Further, a computer readable storage medium stores a computer program which, when executed by a processor, implements the steps of a continuous identity authentication method based on incremental learning and meta learning.
Compared with the prior art, the invention has the following technical effects:
the invention is based on deep learning and meta learning, and no additional limiting condition is added in the acquisition process, thereby fully reflecting the actual mobile phone use condition of the user. And secondly, a two-channel deep learning model 2-GCTN is designed so as to utilize data association information among multiple sensors. The model can acquire data association information among different dimensions and can also remarkably improve the characterization capability of user behaviors under multi-sensor data. Finally, an incremental learning mechanism AMUM based on meta-learning technology is also proposed. The mechanism can update the knowledge learned by the model and the new data in real time, and fully embody the effectiveness of the model. The technical framework MetaAuth of identity authentication based on the touch screen behavior solves the problem in the field of long-term touch screen identity authentication, and secondly designs an incremental learning problem under the scene that an online updating mechanism AMUM based on meta learning is used for continuously carrying out identity authentication for a long time so as to improve the stability of a model.
Drawings
FIG. 1 is a schematic diagram of the sensor data of the present invention.
FIG. 2 is a schematic diagram of an authentication framework based on meta learning and incremental learning according to the present invention.
Fig. 3 is a schematic diagram of a long-term authentication process according to the present invention.
Fig. 4 is a diagram of an experimental view of the effect of the meta-learning mechanism on long-term authentication of a model.
FIG. 5 is a diagram of an experiment using adaptive-MAML as an incremental update mechanism in MetaAuth.
FIG. 6 is a graph of the ability of different models to authenticate in using sensor data.
Detailed Description
Embodiments of the invention herein and experimental procedures will be described in detail as follows:
the first step: firstly, the visualization of the sensor data is carried out, 5 users are randomly extracted, the sensor data is collected in the Taobao app, the total time length is divided into 2 months, and two months can be divided into a front stage, a middle stage and a later stage.
And a second step of: after the features are extracted, the features are visualized in two-dimensional space by a (t-SNE) method, respectively. The 5 users may be represented with similar and varying features between them. As can be clearly seen from fig. 1: at the early stage, the characteristic data of the No. 3, no. 4 and No. 5 are similar, wherein No. 5 is far from the No. 1 comparison. And the original characteristics are kept unchanged by the middle period, but the distances between the No. 1 and the No. 4 are changed more and more. And to the late stage they become further and further apart from each other No. 3, no. 4 and No. 5. It can be seen that the behavior data of the user must change over time, i.e. drift occurs.
And a third step of: in order to solve the problem of performance degradation in long-term identity authentication, an authentication framework MetaAuth based on meta learning and incremental learning is proposed. The overall framework is shown in fig. 2, which contains two important components, not an offline registration component and an online authentication component.
The offline registration stage mainly comprises three steps:
1. and a data acquisition stage: the main approach is to collect data through touch screen behaviors, so long as the data collected by the sensor is used as a positive sample when the screen is lightened, and the data of other users are also used as a negative sample for training.
2. Data preprocessing: in order to effectively extract the characteristics of time series data, a fixed-length sliding window is used to divide the data into a series of time periods where repetition occurs, the time periods are used as basic units for authentication, each time period is divided into disjoint small time periods, namely time sources, and local characteristics can be extracted from the time sources.
3. Continuous collector part: it is desirable to be able to provide some more accurate representation of time-ordered scenes. Therefore, a brand new 2-GCTN model is designed, and the characteristics of each sensor time dimension and the correlation characteristics among different dimensions can be extracted.
While in the online authentication phase: in the process, the authenticator can continuously learn from new data, and the model is updated online by improving the previous method, and the ADWIN method can be used as a drift detector to reduce the resource consumption of the mobile phone end so as to update the model better. Then, a meta learning model is designed, so that the initialization parameters of the model can be learned, and the learning rate of model training can be learned.
At this stage, the model update is performed online using the modified meta learning method AMUM. The device mainly comprises two parts: 1. a drift detector. 2. Improved MAML update policies. Drift detection techniques can be used entirely to improve the efficiency of model updating and his rationality.
In the detection process, the detected time window can be changed according to the time and the change of the data, a drift signal is output according to a prediction result in the window, and if the probability of detecting the error line exceeds a set threshold value, the judgment of drift is given. The next MAML model can improve the efficiency of the model by continuously optimizing the initial parameters of the basic model in task learning.
The specific experimental process comprises the following steps: the first step: experimental data is first acquired, mainly comprising three-dimensional data from sensors of accelerometer, gyroscope and magnetometer.
And a second step of: setting an evaluation index: the accuracy in the scene of identity authentication can be measured by the following indexes, wherein the following indexes mainly are:
1. equal Error Rate (EER): the False Acceptance Rate (FAR) is equal to the value of the False Rejection Rate (FRR), and the equal error rate can be affected by changing the threshold of the prediction score.
2. Error acceptance rate (FAR): defined as the number of illegitimate samples that are received in error and the number of all illegitimate test samples. A higher FAR indicates a higher false separation rate of the intruder sample.
3. False Rejection Rate (FRR): defined as the number of valid samples that were falsely rejected and the number of all valid test samples. A higher FRR indicates a poorer identification of a legitimate user sample.
4. Authentication accuracy variance: a variance measure representing the degree of dispersion of authentication results over a long period of time. It is calculated from the average of the squares of the differences between the EER and EER averages for each authentication. The lower variance reflects the stable predictive ability of the authentication model.
5. And (3) using a memory: representing the amount of memory occupied by model training and updating. The lower memory usage reflects less smartphone resource consumption.
6. Time consumption: representing the time it takes for the model to train and update. Less training time means faster model update speed.
And a third step of: the test was performed. In the experiment, the certification frame 2-GCTN was evaluated by 40 users in the dataset and each user was individually model trained and incrementally tested. The mobile phone corresponding to each user is set as a legal user, and the rest 39 persons are set as illegal users. The data of the positive sample can be selected from each user according to the time sequence, and the data is divided into 20 parts, which respectively correspond to 20 incremental learning tasks. As for the negative samples, random extraction was performed among the remaining 39 users. Each training and test may be defined as a task during the training process. The data block of each next cycle is first used to test the model, and then the training and updating of the model is performed.
The method comprises the following specific steps
1. All processed data are ordered by time and divided into identical data blocks T.
Each data block will be a positive sample data set for the user. Accordingly, there will be T-1 tasks, each task containing two periods of data Di and Di+1.
2. The model is updated using Di as the training set and then tested for its ability using di+1.
3. Step 1 and step 2 are repeated for each user.
In this experiment, the ability of the model, and also the validity of the model, was analyzed by systematic experiments, and also the long-term authentication ability was analyzed.
Firstly, the problem that the model accuracy is reduced under the condition that different models are used by users for a long time is analyzed. In the long-term authentication process, as shown in fig. 3, LSTM, SVM, one-SVM, and 2-GCTN models are used, and the accuracy of these four models is degraded. The reason for this is mainly that the improvement of the environmental complexity causes the behavior of the user using the mobile phone to be changed, and further causes the distribution of the data to be changed. Secondly, models LSTM and 2-GCTN using temporal characteristics initially have better characteristics, but with the change in data distribution, the accuracy drops more than the remaining models.
The effect of the meta-learning mechanism on the long-term authentication of the model is then analyzed. The experimental results are shown in FIG. 4
MetaAuth represents a model update using the AMUM mechanism presented herein. From the figures it can be derived that: when there is no
In the Adaptive-MAML mechanism, the accuracy of the long-term continuous authentication model is reduced and the time to update the model is increased. Because the meta-learning mechanism can obtain better initialized parameters, the model is better to learn. Therefore, the improved meta-learning mechanism can improve the accuracy and the model efficiency of the model under the condition of long-term identity authentication.
The effectiveness of the drift detection mechanism AMUM is discussed next. The new incremental update mechanism AMUM is mainly used for detecting the drift condition of new batch data by using an ADWIN method, and then selecting whether the model is updated or not according to the detected result, so that the MetaAuth structure can update the model at a reasonable frequency and becomes more efficient. The model capabilities of AMUM were compared using conventional AMUM in combination with 2-GCTN, respectively. They are, respectively, retrain-2-GCTN and MetaAuth. In MetaAuth, adaptive-MAML is used as an incremental update mechanism, and only new data is used to update the model. The experimental results are shown in FIG. 5. Experiments show that the EER of MetaAuth is very low, namely, the EER has higher average authentication precision, so that the stability of the model is better. Because the model can utilize information in the historical data through mata-learning. The model can better learn the behavior characteristics of the user. Secondly, the update time of MetaAuth can be obtained in a relatively short time, and the material rate is improved well. It can be concluded that MetaAuth is a more practical and realistic handset scenario.
Finally, long-term authentication capabilities are discussed. The ability of different models to authenticate in using sensor data was compared in experiments. As shown in fig. 6, the proposed model has better accuracy, i.e. minimum EER, with long-term incremental updates, while MetaAuth also has the best stability, compared to the conventional approach. Secondly, a violin diagram with memory usage is adopted, so that I can find that the memory usage of MetaAuth is also the lowest, and the fact that the consumption of resources in the authentication process is the lowest can be shown. Finally, the training time of the MetaAuth is the shortest time in all models, so that the real-time performance of the MetaAuth can be reflected, and the model is the model which is most suitable for a real scene.
In still another embodiment of the present invention, a continuous identity authentication system based on incremental learning and meta learning is provided, which can be used to implement the continuous identity authentication method based on incremental learning and meta learning, and specifically, the system includes:
the data processing judging module is used for collecting sensor data of the mobile phone in the using process, carrying out visual operation on the sensor data and judging whether the sensor data has a drifting phenomenon or not;
the framework building module is used for building an authentication framework MetaAuth comprising an offline registration stage and an online authentication stage and based on meta learning and incremental learning when a drift phenomenon occurs;
the model building module is used for building a meta learning model, acquiring the characteristics of the time dimension of each sensor and the association characteristics among different dimensions of the sensor according to offline registration, learning from the acquired data according to an online authentication stage, and updating the online model;
and the verification module is used for carrying out experimental verification on the updated model to obtain the model capacity.
The division of the modules in the embodiments of the present invention is schematically only one logic function division, and there may be another division manner in actual implementation, and in addition, each functional module in each embodiment of the present invention may be integrated in one processor, or may exist separately and physically, or two or more modules may be integrated in one module. The integrated modules may be implemented in hardware or in software functional modules.
In yet another embodiment of the present invention, a computer device is provided that includes a processor and a memory for storing a computer program including program instructions, the processor for executing the program instructions stored by the computer storage medium. The processor may be a central processing unit (Central Processing Unit, CPU), but may also be other general purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), off-the-shelf Programmable gate arrays (FPGAs) or other Programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc., which are the computational core and control core of the terminal adapted to implement one or more instructions, in particular adapted to load and execute one or more instructions within a computer storage medium to implement the corresponding method flow or corresponding functions; the processor of the embodiment of the invention can be used for the operation of a continuous identity authentication method based on incremental learning and meta learning.
In yet another embodiment of the present invention, a storage medium, specifically a computer readable storage medium (Memory), is a Memory device in a computer device, for storing a program and data. It is understood that the computer readable storage medium herein may include both built-in storage media in a computer device and extended storage media supported by the computer device. The computer-readable storage medium provides a storage space storing an operating system of the terminal. Also stored in the memory space are one or more instructions, which may be one or more computer programs (including program code), adapted to be loaded and executed by the processor. The computer readable storage medium herein may be a high-speed RAM memory or a non-volatile memory (non-volatile memory), such as at least one magnetic disk memory. One or more instructions stored in a computer-readable storage medium may be loaded and executed by a processor to implement the corresponding steps in the above-described embodiments with respect to a continuous identity authentication method based on incremental learning and meta-learning.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Finally, it should be noted that: the above embodiments are only for illustrating the technical aspects of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the above embodiments, it should be understood by those of ordinary skill in the art that: modifications and equivalents may be made to the specific embodiments of the invention without departing from the spirit and scope of the invention, which is intended to be covered by the claims.
Claims (10)
1. The continuous identity authentication method based on incremental learning and meta learning is characterized by comprising the following steps:
collecting sensor data of the mobile phone in the using process, performing visual operation on the sensor data, and judging whether the sensor data has a drifting phenomenon or not;
when drift phenomenon occurs, an authentication framework MetaAuth comprising an offline registration stage and an online authentication stage and based on meta learning and incremental learning is constructed;
establishing a meta learning model, acquiring the characteristics of the time dimension of each sensor and the association characteristics among different dimensions of each sensor according to offline registration, learning from the acquired data according to an online authentication stage, and updating the online model;
and performing experimental verification on the updated model to obtain the model capacity.
2. The continuous identity authentication method based on incremental learning and meta learning according to claim 1, wherein the method is characterized in that sensor data of a mobile phone in a use process is collected, the sensor data is subjected to visual operation, and whether the sensor data has a drift phenomenon is judged:
the method comprises the steps of randomly extracting sensor data when a mobile phone of a user is used in a certain period, dividing the period into a front period, a middle period and a later period, then extracting features of the data, respectively visualizing the features in a two-dimensional space through a t-SNE method after the features are extracted, showing the conditions of similar and changing features among the users, and when the feature data among the users change and the distance is further and further away, drifting occurs.
3. The continuous identity authentication method based on incremental learning and meta learning of claim 1, wherein the offline registration stage:
and (3) data acquisition: data acquisition is carried out by selecting the touch screen behaviors of the user, when the screen is lightened, the data acquired by the sensor is taken as a positive sample, and the data of other users are taken as a negative sample for training;
data preprocessing: firstly, dividing data into a series of time periods with repetition by using a fixed-length sliding window, taking the time periods as a basic unit of authentication, then continuously dividing each time period into disjoint small time periods, namely time sources, and extracting local features from the time sources;
continuous collector part: and establishing a two-channel deep learning model 2-GCTN model, and extracting the characteristics of each sensor time dimension and the associated characteristics among different dimensions.
4. The continuous identity authentication method based on incremental learning and meta learning of claim 1, wherein the online authentication phase:
the on-line model update is carried out at this stage by using an improved meta learning method AMUM, which comprises two parts respectively consisting of: 1. a drift detector; 2. improved MAML update policies; in the detection process, the detected time window can be changed according to the time and the change of the data, a drift signal is output according to a prediction result in the window, and if the probability of detecting the error line exceeds a set threshold value, the judgment of drift is given.
5. The continuous identity authentication method based on incremental learning and meta learning according to claim 4, wherein the specific process is as follows:
firstly, acquiring experimental data comprising three-dimensional data of a sensor from an accelerometer, a gyroscope and a magnetometer;
setting an evaluation index: the accuracy in the scene of identity authentication is measured through indexes, wherein the indexes comprise:
equal error rate: the false acceptance rate is equal to the value of the false rejection rate, and the equal false rate can be affected by changing the threshold of the prediction score;
error acceptance rate: defining the number of illegal samples received by error and the number of all illegal test samples; a high false acceptance rate indicates a high false separation rate of the intruder sample;
false rejection rate: defining the number of valid samples and the number of all valid test samples which are refused by mistake; a high false reject rate indicates poor identification of legitimate user samples;
authentication accuracy variance: a variance measure representing the degree of dispersion of authentication results over a long period of time; it is calculated from the average of the squares of the differences between the equal error rate and the average of the equal error rates for each authentication; the low variance reflects the stable predictive ability of the authentication model;
and (3) using a memory: representing the memory size occupied by model training and updating;
time consumption: representing the time it takes for the model to train and update.
6. A continuous identity authentication method based on incremental learning and meta learning as claimed in claim 3, wherein the test verifies that:
evaluating an authentication framework 2-GCTN through M users in a data set, independently training a model and performing incremental test on each user, setting a mobile phone corresponding to each user as a legal user, and setting the rest M-1 as illegal users; selecting data of a sensor from each user according to the sequence of time for the data of the positive sample, and dividing the data into 20 copies, wherein the 20 copies correspond to the incremental learning tasks respectively; negative samples, randomly extracted among the remaining M-1 users; in the training process, each training and test is defined as a task, and each time the data block of the next period is firstly used for testing the model, and then the training and updating of the model are carried out.
7. The continuous identity authentication method based on incremental learning and meta learning according to claim 6, wherein the specific steps are as follows:
ordering all processed data according to time and dividing the processed data into identical data blocks T;
each data block will be a positive sample data set of the user, there are T-1 tasks, each task containing data D for two periods i D (D) i+1 ；
Using D i Updating the model as a training set, then using D i+1 Testing its ability;
the above steps are repeated for each user.
8. A continuous identity authentication system based on incremental learning and meta learning, comprising:
the data processing judging module is used for collecting sensor data of the mobile phone in the using process, carrying out visual operation on the sensor data and judging whether the sensor data has a drifting phenomenon or not;
the framework building module is used for building an authentication framework MetaAuth comprising an offline registration stage and an online authentication stage and based on meta learning and incremental learning when a drift phenomenon occurs;
the model building module is used for building a meta learning model, acquiring the characteristics of the time dimension of each sensor and the association characteristics among different dimensions of the sensor according to offline registration, learning from the acquired data according to an online authentication stage, and updating the online model;
and the verification module is used for carrying out experimental verification on the updated model to obtain the model capacity.
9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor, when executing the computer program, implements the steps of the continuous identity authentication method based on incremental learning and meta learning as claimed in any one of claims 1 to 7.
10. A computer readable storage medium storing a computer program, characterized in that the computer program when executed by a processor implements the steps of the continuous identity authentication method based on incremental learning and meta learning as claimed in any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310163032.2A CN116028908A (en) | 2023-02-23 | 2023-02-23 | Continuous identity authentication method and related device based on incremental learning and meta learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310163032.2A CN116028908A (en) | 2023-02-23 | 2023-02-23 | Continuous identity authentication method and related device based on incremental learning and meta learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116028908A true CN116028908A (en) | 2023-04-28 |
Family
ID=86072558
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310163032.2A Pending CN116028908A (en) | 2023-02-23 | 2023-02-23 | Continuous identity authentication method and related device based on incremental learning and meta learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116028908A (en) |
-
2023
- 2023-02-23 CN CN202310163032.2A patent/CN116028908A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110009174B (en) | Risk recognition model training method and device and server | |
| CN105590055B (en) | Method and device for identifying user credible behaviors in network interaction system | |
| Traore et al. | Combining mouse and keystroke dynamics biometrics for risk-based authentication in web environments | |
| CN106716382B (en) | The method and system of aggregation multiple utility program behavioural analysis for mobile device behavior | |
| CN110851835A (en) | Image model detection method and device, electronic equipment and storage medium | |
| CN104408341B (en) | Smart phone user identity identifying method based on gyroscope behavioural characteristic | |
| WO2017140222A1 (en) | Modelling method and device for machine learning model | |
| CN107122669B (en) | Method and device for evaluating data leakage risk | |
| CN110442712B (en) | Risk determination method, risk determination device, server and text examination system | |
| CN105337928B (en) | Method for identifying ID, safety protection problem generation method and device | |
| CN107545241A (en) | Neural network model is trained and biopsy method, device and storage medium | |
| US20140351109A1 (en) | Method and apparatus for automatically identifying a fraudulent order | |
| CN109063977B (en) | Non-inductive transaction risk monitoring method and device | |
| CN111159697B (en) | Key detection method and device and electronic equipment | |
| CN112435137B (en) | Cheating information detection method and system based on community mining | |
| CN111275416B (en) | Digital currency abnormal transaction detection method, device, electronic equipment and medium | |
| CN107679626A (en) | Machine learning method, device, system, storage medium and equipment | |
| CN111325619A (en) | Credit card fraud detection model updating method and device based on joint learning | |
| CN102291392A (en) | Hybrid intrusion detection method based on bagging algorithm | |
| CN110084609B (en) | Transaction fraud behavior deep detection method based on characterization learning | |
| WO2022142903A1 (en) | Identity recognition method and apparatus, electronic device, and related product | |
| CN113221104A (en) | User abnormal behavior detection method and user behavior reconstruction model training method | |
| CN113052577A (en) | Method and system for estimating category of virtual address of block chain digital currency | |
| CN112750038B (en) | Transaction risk determination method, device and server | |
| Kumar et al. | Automatic detection of fake profiles in online social networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |