CN117216803B - Intelligent finance-oriented user information protection method and system - Google Patents
Intelligent finance-oriented user information protection method and system Download PDFInfo
- Publication number
- CN117216803B CN117216803B CN202311483288.8A CN202311483288A CN117216803B CN 117216803 B CN117216803 B CN 117216803B CN 202311483288 A CN202311483288 A CN 202311483288A CN 117216803 B CN117216803 B CN 117216803B
- Authority
- CN
- China
- Prior art keywords
- user information
- information protection
- financial product
- session data
- target
- 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
- 238000000034 method Methods 0.000 title claims abstract description 51
- 238000012512 characterization method Methods 0.000 claims description 74
- 238000012544 monitoring process Methods 0.000 claims description 11
- 238000004458 analytical method Methods 0.000 claims description 9
- 238000002372 labelling Methods 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 239000013598 vector Substances 0.000 description 5
- 230000006399 behavior Effects 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 4
- 230000003542 behavioural effect Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000003058 natural language processing Methods 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008929 regeneration Effects 0.000 description 1
- 238000011069 regeneration method Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Abstract
The application provides a user information protection method and system for intelligent finance, which are characterized in that according to first privacy embedded representation information corresponding to target financial product session data and second privacy embedded representation information of each template in X typical user information protection templates, Y typical user information protection templates meeting the matching degree requirement are recommended from the X templates, according to user information protection labels of each template in the Y typical user information protection templates, typical user information protection templates of each user information protection label are analyzed, template matching quantity corresponding to each user information protection label is generated, and a target user information protection strategy corresponding to the target financial product session data is obtained based on the template matching quantity corresponding to each user information protection label. Therefore, the information protection templates meeting the user requirements can be accurately recommended, and the user information protection strategy is determined according to the template matching quantity, so that the user privacy is effectively protected.
Description
Technical Field
The application relates to the technical field of intelligent finance, in particular to a user information protection method and system for intelligent finance.
Background
In the current internet environment, financial service products are frequently used and transacted, and particularly, an intelligent financial service platform, a large amount of user data needs to be collected and processed in order to provide personalized services. However, this also presents challenges for user privacy protection. On the one hand, users are increasingly concerned about how their personal information is handled and protected; on the other hand, laws and regulations have strict requirements for user privacy protection.
In this case, how to effectively protect the privacy of users while satisfying the needs of personalized services becomes an important issue. Conventional approaches typically employ fixed information protection policies, such as encrypting or anonymizing user data. But these methods often ignore the variability between users and fail to meet the specific needs of different users.
Furthermore, while some systems attempt to recommend personalized information protection policies based on the user's behavior or preferences, these systems typically rely on complex user models and large amounts of historical data, resulting in high computational complexity and poor real-time. In addition, these systems also fail to adequately account for the impact of various factors on information protection policy selection, such as privacy preferences of users, service scenarios, and the like.
Therefore, a new method is needed to achieve effective protection of user privacy.
Disclosure of Invention
In view of the foregoing, an object of the present application is to provide a method and a system for protecting user information oriented to intelligent finance.
According to a first aspect of the present application, there is provided a smart finance-oriented user information protection method applied to a smart finance-oriented user information protection system, the method including:
acquiring first privacy embedded characterization information corresponding to target financial product session data in an intelligent financial service page;
recommending Y typical user information protection templates meeting the matching degree requirement from the X typical user information protection templates based on the first privacy embedded representation information and the second privacy embedded representation information of each typical user information protection template in the X typical user information protection templates, wherein each typical user information protection template corresponds to one user information protection tag, Y and X are positive integers, and Y is not more than X;
based on the user information protection labels of each typical user information protection template in the Y typical user information protection templates, analyzing the typical user information protection templates of each user information protection label to generate template matching quantity corresponding to each user information protection label;
And acquiring a target user information protection strategy corresponding to the target financial product session data based on the template matching quantity corresponding to each user information protection label.
In a possible implementation manner of the first aspect, before the obtaining the first privacy embedded characterization information corresponding to the target financial product session data in the smart financial service page, the method further includes:
performing characteristic cross proportion analysis on continuous financial product session data monitored in a session monitoring setting stage, taking financial product session data with characteristic cross proportion not smaller than the set proportion as same session clustering data, and acquiring target financial product session data based on the same session clustering data;
and/or if the generated number of the continuous financial product session data monitored in the set session monitoring stage is greater than the set number, performing feature cross proportion analysis on the continuous financial product session data monitored in the set session monitoring stage, taking the financial product session data with the feature cross proportion not smaller than the set proportion as the same session clustering data, and acquiring the target financial product session data based on the same session clustering data.
In a possible implementation manner of the first aspect, after the obtaining the first privacy embedded representation information corresponding to the target financial product session data in the smart financial service page, the method further includes:
if the typical user information protection template does not exist, generating an extended user information protection tag of the target financial product session data;
based on the coding corresponding information between the first privacy embedded characterization information and the target financial product session data, loading a session source of the target financial product session data and the first privacy embedded characterization information into a first template storage library, wherein the session source of the target financial product session data is used for expressing an index identifier of the target financial product session data;
and loading the session source of the target financial product session data and the extended user information protection tag into a second template storage library based on the coded correspondence information between the extended user information protection tag and the target financial product session data.
In a possible implementation manner of the first aspect, the recommending, based on the first privacy embedded characterizing information and the second privacy embedded characterizing information of each of the X typical user information protection templates, Y typical user information protection templates meeting the matching requirement from the X typical user information protection templates includes:
Based on the first privacy embedded representation information and the second privacy embedded representation information of each of the X typical user information protection templates, recommending the top K typical user information protection templates with the largest feature cross proportion from the X typical user information protection templates based on a first template storage library, wherein K is an integer not less than 1;
determining, for each of the first K typical user information protection templates, a feature cross proportion of the target financial product session data to the typical user information protection template based on the first privacy embedded characterization information and the second privacy embedded characterization information of the typical user information protection template, and if at least one typical user information protection template having a feature cross proportion not less than a set proportion exists in the first K typical user information protection templates, taking the at least one typical user information protection template as the Y typical user information protection templates;
or, for each of the first K typical user information protection templates, determining a feature deviation of the target financial product session data from the typical user information protection template based on the first privacy embedded characterization information and the second privacy embedded characterization information of the typical user information protection template;
And if at least one typical user information protection template with the characteristic deviation degree not larger than the set deviation degree exists in the first K typical user information protection templates, taking the at least one typical user information protection template as the Y typical user information protection templates.
In a possible implementation manner of the first aspect, the method further includes, before recommending, based on the first privacy-embedded feature information and the second privacy-embedded feature information of each of the X typical user information protection templates, the first K typical user information protection templates having the largest feature cross proportion from the X typical user information protection templates based on the first template repository:
acquiring a recommended parameter configuration instruction, wherein the recommended parameter configuration instruction carries the K;
the first privacy embedded representation information and the second privacy embedded representation information of each of the X typical user information protection templates based on the first privacy embedded representation information, and the first K typical user information protection templates with the largest recommended feature cross proportion from the X typical user information protection templates based on the first template repository, include:
If the K is not greater than the X, executing the step of recommending the first K typical user information protection templates with the largest feature cross proportion from the X typical user information protection templates based on a first template storage library based on the first privacy embedded characterization information and the second privacy embedded characterization information of each of the X typical user information protection templates; the method further comprises the steps of:
recommending the X typical user information protection templates based on the first template repository if the K is greater than the X; and acquiring Y typical user information protection templates meeting the matching degree requirement from the X typical user information protection templates.
In a possible implementation manner of the first aspect, after the obtaining the first privacy embedded representation information corresponding to the target financial product session data in the smart financial service page, the method further includes:
if the typical user information protection template meeting the matching degree requirement does not exist, generating an extended user information protection tag of the target financial product session data;
based on the coding corresponding information between the first privacy embedded characterization information and the target financial product session data, loading a session source of the target financial product session data and the first privacy embedded characterization information into a first template storage library, wherein the session source of the target financial product session data is used for expressing the target financial product session data;
And loading the session source of the target financial product session data and the extended user information protection tag into a second template storage library based on the coded correspondence information between the extended user information protection tag and the target financial product session data.
In a possible implementation manner of the first aspect, the generating, based on the user information protection label of each of the Y typical user information protection templates, the template matching number corresponding to each of the Y typical user information protection templates by analyzing the typical user information protection templates of each of the Y typical user information protection templates includes:
based on the user information protection label of each typical user information protection template in the Y typical user information protection templates, associating the typical user information protection templates with the same user information protection label to the same user information protection label, and generating template matching quantity corresponding to each user information protection label;
the obtaining the target user information protection policy corresponding to the target financial product session data based on the template matching number corresponding to each user information protection label includes:
Determining a target user information protection tag corresponding to the target financial product session data based on the template matching quantity corresponding to each user information protection tag;
determining the target user information protection strategy corresponding to the target financial product session data based on the target user information protection label;
wherein, after determining the target user information protection tag corresponding to the target financial product session data based on the template matching number corresponding to each user information protection tag, the method further includes:
based on the coding corresponding information between the first privacy embedded characterization information and the target financial product session data, loading a session source of the target financial product session data and the first privacy embedded characterization information into a first template storage library, wherein the session source of the target financial product session data is used for expressing the target financial product session data;
and loading the session source of the target financial product session data and the target user information protection tag into a second template storage library based on the coding corresponding information between the target user information protection tag and the target financial product session data.
In a possible implementation manner of the first aspect, the determining, based on the number of template matches corresponding to the respective user information protection tags, a target user information protection tag corresponding to the target financial product session data includes:
determining the template matching quantity corresponding to each user information protection tag in at least one user information protection tag, and taking the user information protection tag corresponding to the maximum matching quantity as a target user information protection tag of the target financial product session data;
or determining the template matching quantity corresponding to each user information protection tag in the at least one user information protection tag; if the template matching quantity corresponding to a plurality of user information protection tags is consistent, calculating the average arrangement sequence corresponding to each user information protection tag based on the arrangement sequence of the characteristic cross proportion of the typical user information protection templates corresponding to each user information protection tag in the plurality of user information protection tags; taking the user information protection label corresponding to the maximum average arrangement order as a target user information protection label of the target financial product session data;
Or determining the template matching quantity corresponding to each user information protection tag in the at least one user information protection tag; if the template matching quantity corresponding to a plurality of user information protection tags is consistent, calculating average embedded characterization information corresponding to each user information protection tag based on a typical user information protection template corresponding to each user information protection tag in the plurality of user information protection tags; and determining a target user information protection tag serving as the target financial product session data based on the feature cross proportion between the average embedded characterization information corresponding to each user information protection tag and the first privacy embedded characterization information.
In a possible implementation manner of the first aspect, the determining, based on the target user information protection tag, a target user information protection policy corresponding to the target financial product session data includes:
performing policy condition matching on the target user information protection tag and at least one candidate user information protection policy, and determining a target user information protection policy corresponding to the target financial product session data;
The method further comprises the steps of:
taking a typical user information protection template associated with the target user information protection tag and the target financial product session data as a model learning data sequence, wherein the model learning data sequence carries labeling data of the target user information protection strategy;
acquiring sample privacy embedded characterization information and session account configuration information of each model learning data in the model learning data sequence;
acquiring cluster embedded characterization information of the model learning data sequence;
according to the cluster embedded characterization information of the model learning data sequence, the sample privacy embedded characterization information of each model learning data and the session account configuration information, obtaining a prediction confidence coefficient sequence of each candidate user information protection tag through a user information protection tag prediction network;
training the user information protection tag prediction network based on the target user information protection policy and the prediction confidence sequences of the candidate user information protection tags.
According to a second aspect of the present application, a smart finance-oriented user information protection system is provided, where the smart finance-oriented user information protection system includes a machine-readable storage medium and a processor, where the machine-readable storage medium stores machine-executable instructions, and the processor implements the foregoing smart finance-oriented user information protection method when executing the machine-executable instructions.
According to a third aspect of the present application, there is provided a computer readable storage medium having stored therein computer executable instructions that, when executed, implement the aforementioned smart financial oriented user information protection method.
According to any one of the aspects, in the application, first privacy embedded characterization information corresponding to target financial product session data is obtained from an intelligent financial service page, Y typical user information protection templates meeting the matching degree requirement are recommended from X templates according to the first privacy embedded characterization information and second privacy embedded characterization information of each template in X typical user information protection templates, user information protection tags of each template in the Y typical user information protection templates are analyzed, typical user information protection templates of each user information protection tag are analyzed, template matching quantity corresponding to each user information protection tag is generated, and target user information protection strategies corresponding to the target financial product session data are obtained based on the template matching quantity corresponding to each user information protection tag. Therefore, the information protection templates meeting the user requirements can be accurately recommended, and the user information protection strategy is determined according to the template matching quantity, so that the user privacy is effectively protected.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered limiting in scope, and other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flowchart of a method for protecting user information for intelligent finance according to an embodiment of the present application;
fig. 2 is a schematic component structure diagram of a smart finance-oriented user information protection system for implementing the smart finance-oriented user information protection method according to an embodiment of the present application.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions of the embodiments of the present application will be clearly and completely described below according to the drawings in the embodiments of the present application, and it should be understood that the drawings in the present application are only for the purpose of illustration and description, and are not intended to limit the protection scope of the present application. In addition, it should be understood that the schematic drawings are not drawn to scale. A flowchart, as used in this application, illustrates operations implemented in accordance with some embodiments of the present application. It should be understood that the operations of the flow diagrams may be implemented out of order and that steps without logical context may be performed in reverse order or concurrently. Moreover, one or more other operations may be added to the flow chart or one or more operations may be destroyed from the flow chart as directed by those skilled in the art in light of the present disclosure.
In addition, the described embodiments are only some, but not all, of the embodiments of the present application. The components of the embodiments of the present application, which are generally described and illustrated in the figures herein, may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present application, as provided in the accompanying drawings, is not intended to limit the scope of the application, as claimed, but is merely representative of selected embodiments of the application. All other embodiments, which can be made by a person skilled in the art without making any inventive effort, correspond to the scope of protection of the present application, according to the embodiments of the present application.
Fig. 1 is a flow chart illustrating a smart finance-oriented user information protection method according to an embodiment of the present application, and it should be understood that, in other embodiments, the sequence of part of the steps in the smart finance-oriented user information protection method according to the present embodiment may be shared with each other according to actual needs, or part of the steps may be omitted or maintained. The intelligent finance-oriented user information protection method comprises the following steps of:
step S110, first privacy embedded characterization information corresponding to target financial product session data in the intelligent financial service page is obtained.
For example, assume that an intelligent financial services platform is being processed that provides various financial products such as loans, insurance, investments, etc. The session data generated by the users of the intelligent financial service platform in the using process contains a large amount of private information, so that protecting the privacy of the users is an important task.
For example, when a user browses and queries for an investment product, he may need to enter personal information such as age, income, investment experience, etc., session data that may be converted to first privacy embedded characterization information by some method (e.g., using a deep learning model).
The following is a specific example:
it is assumed that in the wisdom financial services page, when a user browses and queries for an investment product, he may need to enter some personal information such as age, income, investment experience, etc. data. At the same time, the clicking action, browsing time and other data are recorded, and all the data can form the session data of the user.
A pre-trained neural network model can then be used to process the session data. This model may be a Word embedding model (e.g., word2Vec or GloVe) or a more complex model (e.g., BERT). The neural network model is used for converting each piece of session data into a vector, namely the first privacy embedded characterization information. The first privacy embedded characterization information can effectively capture key information in the session data and is represented in the form of numerical values.
For example, for age information, if the user is 30 years old, a vector close to [0.3, 0, 0, ], 0] may be output (here, for example only, the dimensions of the vector may be very high, and the value of each dimension is not necessarily directly related to age). For other types of information such as revenue and investment history, corresponding vectors are also output. Finally, combining all vectors together results in the first privacy-embedded characterizing information for the user session data.
Step S120, recommending Y typical user information protection templates meeting the matching degree requirement from the X typical user information protection templates based on the first privacy embedded representation information and the second privacy embedded representation information of each typical user information protection template in the X typical user information protection templates.
In this embodiment, each typical user information protection template corresponds to a user information protection tag, and Y and X are positive integers, and Y is not greater than X.
A typical user information protection template may be understood as one or more standardized user privacy protection templates that are pre-set, possibly based on specific user attributes or behavioral characteristics, such as age, gender, income, purchasing behavior, browsing behavior, etc., defining for these specific groups how their personal information should be protected.
For example, a typical user information protection template may include the following rules:
for example, assume that there are several typical user information protection templates:
1: the loan product session data for young people includes information about their age, occupation, income, etc.
2: investment product session data for retired elderly people, which may include information about their age, retirement, assets, etc.
3: private banking session data for high equity individuals may include information about their identity, assets, investment preferences, etc.
Then, target financial product session data, such as insurance product session data of a middle-aged person, in the intelligent financial service page is obtained, and the insurance product session data is converted into first privacy embedded characterization information.
Next, a matching representative user information protection template is found based on comparing this first privacy embedded representation information with the second privacy embedded representation information of each representative user information protection template.
That is, each typical user information protection template corresponds to a user information protection label, such as "minors", "adults", "elderly", "high-income", and "invested experience", etc., which is used to represent the user population to which the typical user information protection template is applicable.
Step S130, based on the user information protection label of each typical user information protection template in the Y typical user information protection templates, analyzing the typical user information protection templates of each user information protection label, and generating the template matching quantity corresponding to each user information protection label.
Step S140, obtaining a target user information protection policy corresponding to the target financial product session data based on the template matching number corresponding to each user information protection label.
For example, the target user information protection tag may be subjected to policy condition matching with at least one candidate user information protection policy, so as to determine a target user information protection policy corresponding to the target financial product session data. For example, if the number of "high-income" and "investment-experienced" tags is the greatest, the corresponding target user information protection policy may be to enhance protection of the user's financial status and investment history.
Based on the steps, first privacy embedded characterization information corresponding to target financial product session data is obtained from an intelligent financial service page, Y typical user information protection templates meeting the matching degree requirement are recommended from X templates according to the first privacy embedded characterization information and second privacy embedded characterization information of each template in X typical user information protection templates, the typical user information protection templates of each template in the Y typical user information protection templates are analyzed according to the user information protection tags of each template, template matching quantity corresponding to each user information protection tag is generated, and target user information protection strategies corresponding to the target financial product session data are obtained based on the template matching quantity corresponding to each user information protection tag. Therefore, the information protection templates meeting the user requirements can be accurately recommended, and the user information protection strategy is determined according to the template matching quantity, so that the user privacy is effectively protected.
In a possible embodiment, before step S110, the method further includes: and carrying out characteristic cross proportion analysis on the continuous financial product session data monitored in the session monitoring setting stage, taking the financial product session data with the characteristic cross proportion not smaller than the set proportion as the same session clustering data, and acquiring the target financial product session data based on the same session clustering data.
And/or if the generated number of the continuous financial product session data monitored in the set session monitoring stage is greater than the set number, performing feature cross proportion analysis on the continuous financial product session data monitored in the set session monitoring stage, taking the financial product session data with the feature cross proportion not smaller than the set proportion as the same session clustering data, and acquiring the target financial product session data based on the same session clustering data.
For example, the following is one specific scenario illustration:
it is assumed that an intelligent financial services page is being monitored, on which a user can browse and inquire about various financial products, such as loans, insurance, investments, etc. As the user operates, their behavioral data (e.g., click events, browsing times, etc.) are continuously recorded, forming continuous financial product session data.
Feature cross-scale analysis is performed on the continuous financial product session data monitored during the set session monitoring phase, the purpose of this analysis being to find those session data that contain similar features. For example, if two pieces of session data are related to a user who is about 30 years old, has a stable income, seeks financial advice, then the two pieces of data may have a high degree of feature crossing.
Financial product session data with a feature cross ratio not smaller than the set ratio is regarded as the same session clustering data, and for example, when it is found that the feature cross ratio of some session data reaches a preset threshold (for example, 70%), it can be considered that the data actually belong to the same session clustering. That is, the data reflects a similar class of user behavior or demand.
Acquiring the target financial product session data based on the same session clustering data: once the session clustering is determined, the target financial product session data may be selected therefrom. For example, the most common or typical session data may be selected as the target data.
The above procedure is also applicable to another case where the number of consecutive financial product session data exceeds a preset number (e.g., 1000 pieces) during the set session monitoring phase. In this case, a feature cross-scale analysis is also performed to determine session clustering and obtain target financial product session data therefrom.
In a possible implementation manner, after step S110, the method further includes:
step S111, if the typical user information protection template does not exist, generating an extended user information protection tag of the target financial product session data.
Step S112, based on the encoded correspondence information between the first privacy embedded characterization information and the target financial product session data, loading the session source of the target financial product session data and the first privacy embedded characterization information into a first template repository, where the session source of the target financial product session data is used to express an index identifier of the target financial product session data.
Step S113, based on the coding correspondence information between the extended user information protection tag and the target financial product session data, loads the session source of the target financial product session data and the extended user information protection tag into a second template repository.
For example, the following is one specific scenario illustration:
suppose that in processing session data for a user to query loan products, the session data has been translated into first privacy embedded characterization information. But in the preset typical user information protection template, no template matching this session data is found. In this case, it may be necessary to generate a new extended user information protection tag. For example, a tag such as "middle aged, high income, large loan" may be generated based on information such as age, income, loan amount, etc. of the user.
The session source of the session data (which may be, for example, a particular URL or page ID) may then be stored in the first template store along with the first privacy embedded token information. In this way, an association between session data and its privacy embedded representation can be established.
Thereby, the session source is stored in the second template repository together with the newly generated extended user information protection tag. In this way, this extension tag can be used directly without regeneration when similar session data is encountered in the future.
Suppose a user is processing session data for querying loan products, which has been translated into first privacy embedded characterizing information. There are X preset typical user information protection templates, each having a corresponding second privacy embedded representation information. In one possible implementation, step S120 may include:
step S121, based on the first privacy embedded representation information and the second privacy embedded representation information of each of the X typical user information protection templates, recommending, from the X typical user information protection templates, the first K typical user information protection templates having the largest feature cross proportion based on the first template repository, where K is an integer not less than 1.
Step S122, for each of the first K typical user information protection templates, determining a feature cross ratio of the target financial product session data to the typical user information protection template based on the first privacy embedded feature information and the second privacy embedded feature information of the typical user information protection template, and if at least one typical user information protection template whose feature cross ratio is not less than a set ratio exists in the first K typical user information protection templates, using the at least one typical user information protection template as the Y typical user information protection templates.
First, the first privacy embedded characterization information is compared with the second privacy embedded characterization information of each typical user information protection template, and the first K typical user information protection templates with the largest feature cross proportion are found out. For example, if there are 10 typical user information protection templates, the first 3 typical user information protection templates with the largest feature intersection proportion may be selected. And further calculating the characteristic cross proportion of the target session data and each typical user information protection template aiming at the K typical user information protection templates. If the feature cross proportion of at least one of the K typical subscriber information protection templates reaches a preset threshold (e.g., 70%), then the K typical subscriber information protection templates are selected as Y typical subscriber information protection templates.
Feature cross-scaling is an important indicator of similarity between the target session data and the typical user information protection template. The calculation generally involves two steps: feature extraction and feature matching.
The following is one possible calculation method:
first, significant features need to be extracted from the target session data and the typical user information protection template. These characteristics may include the user's age, gender, income, etc., as well as behavioral data of the user, such as click events, browsing times, etc. In this process, some machine learning or deep learning techniques may be used, such as Word-embedding models (Word 2Vec, gloVe, etc.), natural Language Processing (NLP) techniques, etc.
The characteristics of the target session data are then matched with the characteristics of each typical user information protection template. A feature is considered intersected if it exists in both the target session data and the template. For example, if the target session data indicates that the user is about 30 years old, high income, seeking loan products, and the typical user information protection template also indicates that the user is about 30 years old, high income, seeking loan products, then there is a high feature crossover between the two.
Feature intersection ratio may be defined as the ratio of the number of intersecting features to the total number of features. For example, if the target session data and template have 10 features in total, 7 of which are intersected, then the feature intersection ratio is 70%.
The above is just one possible calculation method, and the actual calculation process may be adjusted according to the specific application scenario and requirement.
Or, in step S123, for each of the first K typical user information protection templates, a feature deviation degree of the target financial product session data from the typical user information protection template is determined based on the first privacy embedded characterization information and the second privacy embedded characterization information of the typical user information protection template.
And step S124, if at least one typical user information protection template with the characteristic deviation degree not larger than the set deviation degree exists in the first K typical user information protection templates, the at least one typical user information protection template is used as the Y typical user information protection templates.
For example, another method is to calculate the degree of feature deviation of the target session data from each typical user information protection template, that is, the degree of difference therebetween. If the feature deviation of at least one of the K typical subscriber information protection templates does not exceed a preset threshold (e.g., 30%), then the K typical subscriber information protection templates are selected as Y typical subscriber information protection templates.
In a possible embodiment, before step S120, the method further includes:
and acquiring a recommended parameter configuration instruction, wherein the recommended parameter configuration instruction carries the K.
Step S120 may include:
and if the K is not greater than the X, executing the step of recommending the first K typical user information protection templates with the largest feature cross proportion from the X typical user information protection templates based on a first template storage library based on the first privacy embedded characterization information and the second privacy embedded characterization information of each of the X typical user information protection templates. If the K is greater than the X, recommending the X typical user information protection templates based on the first template repository. And acquiring Y typical user information protection templates meeting the matching degree requirement from the X typical user information protection templates.
In a possible implementation manner, after step S110, if there is no typical user information protection template meeting the matching requirement, an extended user information protection tag of the target financial product session data is generated, and based on the encoded correspondence information between the first privacy embedded characterization information and the target financial product session data, a session source of the target financial product session data and the first privacy embedded characterization information are loaded into a first template repository, where the session source of the target financial product session data is used to express the target financial product session data. Then, based on the code correspondence information between the extended user information protection tag and the target financial product session data, the session source of the target financial product session data and the extended user information protection tag are loaded into a second template repository.
In a possible implementation manner, in step S120, based on the user information protection label of each of the Y typical user information protection templates, the typical user information protection templates with the same user information protection label are associated to the same user information protection label, and the template matching number corresponding to each of the Y typical user information protection templates is generated.
In step S140, a target user information protection tag corresponding to the target financial product session data is determined based on the template matching number corresponding to each user information protection tag, and a target user information protection policy corresponding to the target financial product session data is determined based on the target user information protection tag.
After step S130, based on the encoded correspondence information between the first privacy embedded characterization information and the target financial product session data, a session source of the target financial product session data and the first privacy embedded characterization information are loaded into a first template repository, where the session source of the target financial product session data is used to express the target financial product session data. Then, based on the coding corresponding information between the target user information protection tag and the target financial product session data, the session source of the target financial product session data and the target user information protection tag are loaded into a second template storage library.
In one possible implementation, step S130 may include: and determining the template matching quantity corresponding to each user information protection tag in at least one user information protection tag, and taking the user information protection tag corresponding to the maximum matching quantity as a target user information protection tag of the target financial product session data. Or determining the template matching quantity corresponding to each user information protection label in the at least one user information protection label. If the template matching quantity corresponding to a plurality of user information protection tags is consistent, calculating the average arrangement sequence corresponding to each user information protection tag based on the arrangement sequence of the characteristic cross proportion of the typical user information protection templates corresponding to each user information protection tag in the plurality of user information protection tags. And taking the user information protection label corresponding to the maximum average arrangement order as the target user information protection label of the target financial product session data. Or determining the template matching quantity corresponding to each user information protection label in the at least one user information protection label. If the template matching quantity corresponding to the plurality of user information protection tags is consistent, calculating average embedded characterization information corresponding to each user information protection tag based on the typical user information protection template corresponding to each user information protection tag in the plurality of user information protection tags. And determining a target user information protection tag serving as the target financial product session data based on the feature cross proportion between the average embedded characterization information corresponding to each user information protection tag and the first privacy embedded characterization information.
In one possible implementation manner, determining, based on the target user information protection tag, a target user information protection policy corresponding to the target financial product session data includes: and carrying out strategy condition matching on the target user information protection tag and at least one candidate user information protection strategy, and determining the target user information protection strategy corresponding to the target financial product session data.
On the basis of the above description, the embodiment of the application may further include:
step S101, a typical user information protection template associated with the target user information protection tag and the target financial product session data are used as a model learning data sequence, wherein the model learning data sequence carries labeling data of the target user information protection strategy.
Step S102, sample privacy embedded characterization information and session account configuration information of each model learning data in the model learning data sequence are obtained.
Step S103, cluster embedded characterization information of the model learning data sequence is obtained.
Step S104, according to the cluster embedded characterization information of the model learning data sequence, the sample privacy embedded characterization information of each model learning data and the session account configuration information, a prediction confidence sequence of each candidate user information protection tag is obtained through a user information protection tag prediction network.
Step S105, training the user information protection tag prediction network based on the target user information protection policy and the prediction confidence sequences of the candidate user information protection tags.
Fig. 2 schematically illustrates a smart finance-oriented user information protection system 100 that may be used to implement various embodiments described herein.
For one embodiment, FIG. 2 illustrates a smart-finance oriented user information protection system 100, the smart-finance oriented user information protection system 100 having one or more processors 102, a control module (chipset) 104 coupled to one or more of the processor(s) 102, a memory 106 coupled to the control module 104, a non-volatile memory (NVY)/storage device 108 coupled to the control module 104, one or more input/output devices 110 coupled to the control module 104, and a network interface 112 coupled to the control module 104.
The processor 102 may include one or more single-core or multi-core processors, and the processor 102 may include any combination of general-purpose or special-purpose processors (e.g., graphics processors, application processors, baseband processors, etc.). In an alternative embodiment, the smart finance-oriented user information protection system 100 can be used as a server device such as a gateway in the embodiments of the present application.
Fig. 2 schematically illustrates a smart finance-oriented user information protection system 100 that may be used to implement various embodiments described herein.
For one embodiment, FIG. 2 illustrates a smart-finance oriented user information protection system 100, the smart-finance oriented user information protection system 100 having one or more processors 102, a control module (chipset) 104 coupled to one or more of the processor(s) 102, a memory 106 coupled to the control module 104, a non-volatile memory (NVM)/storage device 108 coupled to the control module 104, one or more input/output devices 110 coupled to the control module 104, and a network interface 112 coupled to the control module 104.
The processor 102 may include one or more single-core or multi-core processors, and the processor 102 may include any combination of general-purpose or special-purpose processors (e.g., graphics processors, application processors, baseband processors, etc.). In an alternative embodiment, the smart finance-oriented user information protection system 100 can be used as a server device such as a gateway in the embodiments of the present application.
In an alternative embodiment, the smart finance-oriented user information protection system 100 may include one or more computer-readable media (e.g., memory 106 or NVM/storage 108) having instructions 114 and one or more processors 102, in combination with the one or more computer-readable media, configured to execute the instructions 114 to implement the modules to perform the actions described in this disclosure.
For one embodiment, the control module 104 may include any suitable interface controller to provide any suitable interface to one or more of the processor(s) 102 and/or any suitable device or component in communication with the control module 104.
The control module 104 may include a memory controller module to provide an interface to the memory 106. The memory controller modules may be hardware modules, software modules, and/or firmware modules.
Memory 106 may be used to load and store data and/or instructions 114 for, for example, smart finance-oriented user information protection system 100. For one embodiment, memory 106 may comprise any suitable volatile memory, such as, for example, a suitable DRAM. In an alternative embodiment, memory 106 may comprise a double data rate type four synchronous dynamic random access memory (DDR 4 SDRAM).
For one embodiment, control module 104 may include one or more input/output controllers to provide interfaces to NVM/storage 108 and input/output device(s) 110.
For example, NVM/storage 108 may be used to store data and/or instructions 114. NVM/storage 108 may include any suitable nonvolatile memory (e.g., flash memory) and/or may include any suitable nonvolatile storage(s) (e.g., one or more Hard Disk Drives (HDDs), one or more Compact Disc (CD) drives, and/or one or more Digital Versatile Disc (DVD) drives).
NVM/storage 108 may include storage resources that are physically part of the device on which smart-finance oriented user information protection system 100 is installed, or which may be accessible by the device, but may not be necessary as part of the device. For example, NVM/storage 108 may be accessed via input/output device(s) 110 according to a network.
Input/output device(s) 110 may provide an interface for smart financial oriented user information protection system 100 to communicate with any other suitable device, input/output device 110 may include a communication component, pinyin component, sensor component, and the like. The network interface 112 may provide an interface for the smart-finance oriented user information protection system 100 to communicate in accordance with one or more networks, the smart-finance oriented user information protection system 100 may communicate wirelessly with one or more components of a wireless network in accordance with any of one or more wireless network standards and/or protocols, such as accessing a wireless network in accordance with a communication standard, or a combination thereof.
For one embodiment, one or more of the processor(s) 102 may be loaded with logic of one or more controllers (e.g., memory controller modules) of the control module 104. For one embodiment, one or more of the processor(s) 102 may be loaded together with logic of one or more controllers of the control module 104 to form a system level load. For one embodiment, one or more of the processor(s) 102 may be integrated on the same mold as logic of one or more controllers of the control module 104. For one embodiment, one or more of the processor(s) 102 may be integrated on the same die with logic of one or more controllers of the control module 104 to form a system on chip (SoC).
In various embodiments, the smart finance-oriented user information protection system 100 may be, but is not limited to: a server, a desktop computing device, or a mobile computing device (e.g., a laptop computing device, a handheld computing device, a tablet, a netbook, etc.), among other terminal devices. In various embodiments, the smart finance-oriented user information protection system 100 may have more or fewer components and/or different architectures. For example, in one alternative embodiment, smart financial oriented user information protection system 100 includes one or more cameras, a keyboard, a Liquid Crystal Display (LCD) screen (including a touch screen display), a non-volatile memory port, multiple antennas, a graphics chip, an Application Specific Integrated Circuit (ASIC), and speakers.
The foregoing has outlined rather broadly the more detailed description of embodiments of the present application, wherein specific examples are provided herein to illustrate the principles and embodiments of the present application, the above examples being provided solely to assist in the understanding of the methods of the present application and the core ideas thereof; meanwhile, as those skilled in the art will have modifications in the specific embodiments and application scope in accordance with the ideas of the present application, the present description should not be construed as limiting the present application in view of the above.
Claims (8)
1. The intelligent finance-oriented user information protection method is characterized by being applied to an intelligent finance-oriented user information protection system, and comprises the following steps:
acquiring first privacy embedded characterization information corresponding to target financial product session data in an intelligent financial service page;
recommending Y typical user information protection templates meeting the matching degree requirement from the X typical user information protection templates based on the first privacy embedded representation information and the second privacy embedded representation information of each typical user information protection template in the X typical user information protection templates, wherein each typical user information protection template corresponds to one user information protection tag, Y and X are positive integers, and Y is not more than X;
based on the user information protection labels of each typical user information protection template in the Y typical user information protection templates, analyzing the typical user information protection templates of each user information protection label to generate template matching quantity corresponding to each user information protection label;
acquiring a target user information protection strategy corresponding to the target financial product session data based on the template matching quantity corresponding to each user information protection label;
The user information protection label based on each typical user information protection template in the Y typical user information protection templates analyzes the typical user information protection templates of each user information protection label, and generates template matching quantity corresponding to each user information protection label, including:
based on the user information protection label of each typical user information protection template in the Y typical user information protection templates, associating the typical user information protection templates with the same user information protection label to the same user information protection label, and generating template matching quantity corresponding to each user information protection label;
the obtaining the target user information protection policy corresponding to the target financial product session data based on the template matching number corresponding to each user information protection label includes:
determining a target user information protection tag corresponding to the target financial product session data based on the template matching quantity corresponding to each user information protection tag;
determining the target user information protection strategy corresponding to the target financial product session data based on the target user information protection label;
Wherein, after determining the target user information protection tag corresponding to the target financial product session data based on the template matching number corresponding to each user information protection tag, the method further includes:
based on the coding corresponding information between the first privacy embedded characterization information and the target financial product session data, loading a session source of the target financial product session data and the first privacy embedded characterization information into a first template storage library, wherein the session source of the target financial product session data is used for expressing the target financial product session data;
based on the coding corresponding information between the target user information protection tag and the target financial product session data, loading a session source of the target financial product session data and the target user information protection tag into a second template storage library;
the determining the target user information protection label corresponding to the target financial product session data based on the template matching quantity corresponding to each user information protection label includes:
determining the template matching quantity corresponding to each user information protection tag in at least one user information protection tag, and taking the user information protection tag corresponding to the maximum matching quantity as a target user information protection tag of the target financial product session data;
Or determining the template matching quantity corresponding to each user information protection tag in the at least one user information protection tag; if the template matching quantity corresponding to a plurality of user information protection tags is consistent, calculating the average arrangement sequence corresponding to each user information protection tag based on the arrangement sequence of the characteristic cross proportion of the typical user information protection templates corresponding to each user information protection tag in the plurality of user information protection tags; taking the user information protection label corresponding to the maximum average arrangement order as a target user information protection label of the target financial product session data;
or determining the template matching quantity corresponding to each user information protection tag in the at least one user information protection tag; if the template matching quantity corresponding to a plurality of user information protection tags is consistent, calculating average embedded characterization information corresponding to each user information protection tag based on a typical user information protection template corresponding to each user information protection tag in the plurality of user information protection tags; and determining a target user information protection tag serving as the target financial product session data based on the feature cross proportion between the average embedded characterization information corresponding to each user information protection tag and the first privacy embedded characterization information.
2. The smart finance-oriented user information protection method of claim 1, wherein before obtaining the first privacy embedded representation information corresponding to the target financial product session data in the smart financial service page, the method further comprises:
performing characteristic cross proportion analysis on continuous financial product session data monitored in a session monitoring setting stage, taking financial product session data with characteristic cross proportion not smaller than the set proportion as same session clustering data, and acquiring target financial product session data based on the same session clustering data;
and/or if the generated number of the continuous financial product session data monitored in the set session monitoring stage is greater than the set number, performing feature cross proportion analysis on the continuous financial product session data monitored in the set session monitoring stage, taking the financial product session data with the feature cross proportion not smaller than the set proportion as the same session clustering data, and acquiring the target financial product session data based on the same session clustering data.
3. The smart financial oriented user information protection method of claim 1, wherein after obtaining the first privacy embedded representation information corresponding to the target financial product session data in the smart financial service page, the method further comprises:
If the typical user information protection template does not exist, generating an extended user information protection tag of the target financial product session data;
based on the coding corresponding information between the first privacy embedded characterization information and the target financial product session data, loading a session source of the target financial product session data and the first privacy embedded characterization information into a first template storage library, wherein the session source of the target financial product session data is used for expressing an index identifier of the target financial product session data;
and loading the session source of the target financial product session data and the extended user information protection tag into a second template storage library based on the coded correspondence information between the extended user information protection tag and the target financial product session data.
4. The intelligent finance oriented user information protection method according to claim 1, wherein the recommending Y typical user information protection templates meeting the matching degree requirement from the X typical user information protection templates based on the first privacy embedded characterization information and the second privacy embedded characterization information of each of the X typical user information protection templates includes:
Based on the first privacy embedded representation information and the second privacy embedded representation information of each of the X typical user information protection templates, recommending the top K typical user information protection templates with the largest feature cross proportion from the X typical user information protection templates based on a first template storage library, wherein K is an integer not less than 1;
determining, for each of the first K typical user information protection templates, a feature cross proportion of the target financial product session data to the typical user information protection template based on the first privacy embedded characterization information and the second privacy embedded characterization information of the typical user information protection template, and if at least one typical user information protection template having a feature cross proportion not less than a set proportion exists in the first K typical user information protection templates, taking the at least one typical user information protection template as the Y typical user information protection templates;
or, for each of the first K typical user information protection templates, determining a feature deviation of the target financial product session data from the typical user information protection template based on the first privacy embedded characterization information and the second privacy embedded characterization information of the typical user information protection template;
And if at least one typical user information protection template with the characteristic deviation degree not larger than the set deviation degree exists in the first K typical user information protection templates, taking the at least one typical user information protection template as the Y typical user information protection templates.
5. The intelligent finance oriented user information protection method of claim 4, wherein the first privacy embedded representation information and the second privacy embedded representation information of each of the X typical user information protection templates are based on a first template repository recommending top K typical user information protection templates with the largest feature cross-over ratio from the X typical user information protection templates, the method further comprising:
acquiring a recommended parameter configuration instruction, wherein the recommended parameter configuration instruction carries the K;
the first privacy embedded representation information and the second privacy embedded representation information of each of the X typical user information protection templates based on the first privacy embedded representation information, and the first K typical user information protection templates with the largest recommended feature cross proportion from the X typical user information protection templates based on the first template repository, include:
If the K is not greater than the X, executing the step of recommending the first K typical user information protection templates with the largest feature cross proportion from the X typical user information protection templates based on a first template storage library based on the first privacy embedded characterization information and the second privacy embedded characterization information of each of the X typical user information protection templates; the method further comprises the steps of:
recommending the X typical user information protection templates based on the first template repository if the K is greater than the X; and acquiring Y typical user information protection templates meeting the matching degree requirement from the X typical user information protection templates.
6. The smart financial oriented user information protection method of claim 1, wherein after obtaining the first privacy embedded representation information corresponding to the target financial product session data in the smart financial service page, the method further comprises:
if the typical user information protection template meeting the matching degree requirement does not exist, generating an extended user information protection tag of the target financial product session data;
Based on the coding corresponding information between the first privacy embedded characterization information and the target financial product session data, loading a session source of the target financial product session data and the first privacy embedded characterization information into a first template storage library, wherein the session source of the target financial product session data is used for expressing the target financial product session data;
and loading the session source of the target financial product session data and the extended user information protection tag into a second template storage library based on the coded correspondence information between the extended user information protection tag and the target financial product session data.
7. The intelligent finance-oriented user information protection method according to claim 1, wherein the determining a target user information protection policy corresponding to the target financial product session data based on the target user information protection tag includes:
performing policy condition matching on the target user information protection tag and at least one candidate user information protection policy, and determining a target user information protection policy corresponding to the target financial product session data;
the method further comprises the steps of:
Taking a typical user information protection template associated with the target user information protection tag and the target financial product session data as a model learning data sequence, wherein the model learning data sequence carries labeling data of the target user information protection strategy;
acquiring sample privacy embedded characterization information and session account configuration information of each model learning data in the model learning data sequence;
acquiring cluster embedded characterization information of the model learning data sequence;
according to the cluster embedded characterization information of the model learning data sequence, the sample privacy embedded characterization information of each model learning data and the session account configuration information, obtaining a prediction confidence coefficient sequence of each candidate user information protection tag through a user information protection tag prediction network;
training the user information protection tag prediction network based on the target user information protection policy and the prediction confidence sequences of the candidate user information protection tags.
8. A smart finance oriented user information protection system comprising a processor and a computer readable storage medium storing machine executable instructions that when executed by the processor implement the smart finance oriented user information protection method of any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311483288.8A CN117216803B (en) | 2023-11-09 | 2023-11-09 | Intelligent finance-oriented user information protection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311483288.8A CN117216803B (en) | 2023-11-09 | 2023-11-09 | Intelligent finance-oriented user information protection method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117216803A CN117216803A (en) | 2023-12-12 |
| CN117216803B true CN117216803B (en) | 2024-02-09 |
Family
ID=89044813
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311483288.8A Active CN117216803B (en) | 2023-11-09 | 2023-11-09 | Intelligent finance-oriented user information protection method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117216803B (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103455737A (en) * | 2012-05-28 | 2013-12-18 | 百度在线网络技术（北京）有限公司 | User information protection method and device |
| CN103596172A (en) * | 2013-11-25 | 2014-02-19 | 中国联合网络通信集团有限公司 | User information protection method, device and system |
| CN105809464A (en) * | 2014-12-31 | 2016-07-27 | 中国电信股份有限公司 | Method and device for information delivery |
| CN109033846A (en) * | 2018-06-08 | 2018-12-18 | 浙江捷尚人工智能研究发展有限公司 | Privacy of user guard method and system |
| CN110287408A (en) * | 2019-05-23 | 2019-09-27 | 上海拍拍贷金融信息服务有限公司 | A kind of method and system customizing push financial product |
| CN112184429A (en) * | 2020-11-06 | 2021-01-05 | 上海优扬新媒信息技术有限公司 | User information processing method and block chain link point |
| CN112765670A (en) * | 2021-02-03 | 2021-05-07 | 浙江机电职业技术学院 | User information service privacy protection method and system based on identification replacement |
| CN113032668A (en) * | 2021-03-03 | 2021-06-25 | 深圳市七号网络科技有限公司 | Product recommendation method, device and equipment based on user portrait and storage medium |
| CN114549174A (en) * | 2022-02-16 | 2022-05-27 | 中国工商银行股份有限公司 | User behavior prediction method and device, computer equipment and storage medium |
| CN115002200A (en) * | 2022-05-31 | 2022-09-02 | 平安银行股份有限公司 | User portrait based message pushing method, device, equipment and storage medium |
| CN116155538A (en) * | 2022-12-06 | 2023-05-23 | 零束科技有限公司 | Privacy protection method, device, electronic equipment and computer storage medium |
| CN116993512A (en) * | 2023-08-22 | 2023-11-03 | 平安银行股份有限公司 | Recommendation method, system, equipment and storage medium for financial products |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7774715B1 (en) * | 2000-06-23 | 2010-08-10 | Ecomsystems, Inc. | System and method for computer-created advertisements |
-
2023
- 2023-11-09 CN CN202311483288.8A patent/CN117216803B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103455737A (en) * | 2012-05-28 | 2013-12-18 | 百度在线网络技术（北京）有限公司 | User information protection method and device |
| CN103596172A (en) * | 2013-11-25 | 2014-02-19 | 中国联合网络通信集团有限公司 | User information protection method, device and system |
| CN105809464A (en) * | 2014-12-31 | 2016-07-27 | 中国电信股份有限公司 | Method and device for information delivery |
| CN109033846A (en) * | 2018-06-08 | 2018-12-18 | 浙江捷尚人工智能研究发展有限公司 | Privacy of user guard method and system |
| CN110287408A (en) * | 2019-05-23 | 2019-09-27 | 上海拍拍贷金融信息服务有限公司 | A kind of method and system customizing push financial product |
| CN112184429A (en) * | 2020-11-06 | 2021-01-05 | 上海优扬新媒信息技术有限公司 | User information processing method and block chain link point |
| CN112765670A (en) * | 2021-02-03 | 2021-05-07 | 浙江机电职业技术学院 | User information service privacy protection method and system based on identification replacement |
| CN113032668A (en) * | 2021-03-03 | 2021-06-25 | 深圳市七号网络科技有限公司 | Product recommendation method, device and equipment based on user portrait and storage medium |
| CN114549174A (en) * | 2022-02-16 | 2022-05-27 | 中国工商银行股份有限公司 | User behavior prediction method and device, computer equipment and storage medium |
| CN115002200A (en) * | 2022-05-31 | 2022-09-02 | 平安银行股份有限公司 | User portrait based message pushing method, device, equipment and storage medium |
| CN116155538A (en) * | 2022-12-06 | 2023-05-23 | 零束科技有限公司 | Privacy protection method, device, electronic equipment and computer storage medium |
| CN116993512A (en) * | 2023-08-22 | 2023-11-03 | 平安银行股份有限公司 | Recommendation method, system, equipment and storage medium for financial products |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117216803A (en) | 2023-12-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2021004333A1 (en) | Knowledge graph-based event processing method and apparatus, device, and storage medium | |
| US11531987B2 (en) | User profiling based on transaction data associated with a user | |
| CN112863683B (en) | Medical record quality control method and device based on artificial intelligence, computer equipment and storage medium | |
| CN114298417A (en) | Anti-fraud risk assessment method, anti-fraud risk training method, anti-fraud risk assessment device, anti-fraud risk training device and readable storage medium | |
| CN114612759B (en) | Video processing method, video query method, model training method and model training device | |
| CN110688974A (en) | Identity recognition method and device | |
| CN113177700B (en) | Risk assessment method, system, electronic equipment and storage medium | |
| CN112231592A (en) | Network community discovery method, device, equipment and storage medium based on graph | |
| US20230004979A1 (en) | Abnormal behavior detection method and apparatus, electronic device, and computer-readable storage medium | |
| CN111538794A (en) | Data fusion method, device and equipment | |
| CN115544560A (en) | Desensitization method and device for sensitive information, computer equipment and storage medium | |
| CN115146068A (en) | Method, device and equipment for extracting relation triples and storage medium | |
| CN116542673B (en) | Fraud identification method and system applied to machine learning | |
| CN117216803B (en) | Intelligent finance-oriented user information protection method and system | |
| CN108768742B (en) | Network construction method and device, electronic equipment and storage medium | |
| CN115983982A (en) | Credit risk identification method, credit risk identification device, credit risk identification equipment and computer readable storage medium | |
| CN111316259A (en) | System and method for dynamic synthesis and transient clustering of semantic attributes for feedback and adjudication | |
| CN115099875A (en) | Data classification method based on decision tree model and related equipment | |
| CN110837596B (en) | Intelligent recommendation method and device, computer equipment and storage medium | |
| CN113807920A (en) | Artificial intelligence based product recommendation method, device, equipment and storage medium | |
| Zhang et al. | A ResNet-LSTM Based Credit Scoring Approach for Imbalanced Data | |
| CN113239215A (en) | Multimedia resource classification method and device, electronic equipment and storage medium | |
| US11971901B1 (en) | Systems for encoding data transforms by intent | |
| CN112905987B (en) | Account identification method, device, server and storage medium | |
| CN111190967B (en) | User multidimensional data processing method and device and electronic equipment |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |