CN116976661A - Rule generation method, device, computer equipment and storage medium - Google Patents
Rule generation method, device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN116976661A CN116976661A CN202211681274.2A CN202211681274A CN116976661A CN 116976661 A CN116976661 A CN 116976661A CN 202211681274 A CN202211681274 A CN 202211681274A CN 116976661 A CN116976661 A CN 116976661A
- Authority
- CN
- China
- Prior art keywords
- feature
- characteristic
- target
- value
- screening
- 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
- 238000000034 method Methods 0.000 title claims abstract description 67
- 238000012216 screening Methods 0.000 claims abstract description 212
- 238000012954 risk control Methods 0.000 claims abstract description 192
- 238000012795 verification Methods 0.000 claims abstract description 74
- 230000000694 effects Effects 0.000 claims abstract description 54
- 238000012545 processing Methods 0.000 claims abstract description 47
- 230000000875 corresponding effect Effects 0.000 claims description 250
- 238000004590 computer program Methods 0.000 claims description 23
- 230000002596 correlated effect Effects 0.000 claims description 6
- 238000004891 communication Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000001914 filtration Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/60—Protecting data
- G06F21/64—Protecting data integrity, e.g. using checksums, certificates or signatures
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Computer Security & Cryptography (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- Educational Administration (AREA)
- Quality & Reliability (AREA)
- Health & Medical Sciences (AREA)
- Bioethics (AREA)
- General Health & Medical Sciences (AREA)
- Computer Hardware Design (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses a rule generation method, a device, computer equipment and a storage medium, wherein the method comprises the following steps: acquiring at least one feature set, wherein one feature set comprises feature values of different sample objects under one feature dimension; based on a characteristic value screening rule and a value range of a characteristic value in any characteristic set, carrying out characteristic value screening processing on the characteristic set to obtain a screening characteristic range of the characteristic set and an indication parameter of a screening effect corresponding to the screening characteristic range; according to the indication parameters corresponding to each feature set, determining a target feature set from the feature sets of which the indication parameters meet parameter thresholds, after the feature values of the target feature sets are subjected to compliance verification, indicating that the feature values in the corresponding target feature sets meet the requirements of a target wind control scene by the presence of a verification result, and generating a risk control rule of the target wind control scene based on the obtained feature values and the corresponding feature dimensions of the target feature sets.
Description
Technical Field
The present application relates to the field of computer technologies, and in particular, to a rule generating method, a rule generating device, a computer device, and a storage medium.
Background
With the development of computer technology, more and more people join in the internet, so that each internet industry has a huge number of business objects. In order to ensure the operation effect of the internet service of the provider in each internet industry and further provide high-quality service for the service object, the risk control rule may be adopted to perform relevant constraint on the execution condition of the internet service. The risk control rule generated by the current rule generation method has an undesirable application effect in the related application scene. Therefore, how to effectively generate risk control rules under specific wind control scenes becomes a current research hotspot.
Disclosure of Invention
The embodiment of the application provides a rule generation method, a rule generation device, computer equipment and a storage medium, which can effectively generate a risk control rule under a target wind control scene.
In one aspect, an embodiment of the present application provides a rule generating method, including:
acquiring at least one feature set, wherein one feature set corresponds to one feature dimension, and any feature set contains feature values of different sample objects under the corresponding feature dimension; the characteristic value of any sample object is acquired under a target wind control scene;
Based on a characteristic value screening rule, combining a value range of a characteristic value in any characteristic set, and performing characteristic value screening processing on the any characteristic set to obtain a screening characteristic range of the any characteristic set and an indication parameter of a screening effect corresponding to the screening characteristic range of the any characteristic set;
determining a target feature set from feature sets, corresponding to the indication parameters of the screening effect, of which the corresponding indication parameters meet parameter thresholds, and carrying out compliance verification on feature values under the obtained target feature set in the target wind control scene to obtain at least one verification result;
and when the obtained at least one verification result indicates that the characteristic value under the corresponding target characteristic set meets the requirement of the target wind control scene by the reference verification result, generating a risk control rule of the target wind control scene based on the characteristic dimension corresponding to the obtained target characteristic set and the characteristic value under the corresponding target characteristic set.
In one aspect, an embodiment of the present application provides a rule generating apparatus, including:
the device comprises an acquisition unit, a storage unit and a processing unit, wherein the acquisition unit is used for acquiring at least one feature set, one feature set corresponds to one feature dimension, and any feature set contains feature values of different sample objects under corresponding feature dimensions; the characteristic value of any sample object is acquired under a target wind control scene;
The feature set processing unit is used for carrying out feature value screening processing on any feature set based on a feature value screening rule and combining a value range of a feature value in the any feature set to obtain a screening feature range of the any feature set and an indication parameter of a screening effect corresponding to the screening feature range of the any feature set;
the verification unit is used for determining a target feature set from the feature set of which the corresponding indication parameter meets a parameter threshold according to the indication parameter of the screening effect corresponding to each feature set, and carrying out compliance verification on the feature value under the obtained target feature set under the target wind control scene to obtain at least one verification result;
the generation unit is used for generating a risk control rule of the target wind control scene based on the feature dimension corresponding to the obtained target feature set and the feature value under the corresponding target feature set when the reference check result indicates that the feature value under the corresponding target feature set meets the requirement of the target wind control scene.
In one aspect, an embodiment of the present application provides a computer device, including:
a processor for implementing one or more computer programs;
A computer storage medium storing one or more computer programs adapted to be loaded by the processor and to implement a method as in the first aspect.
In one aspect, embodiments of the present application provide a storage medium storing one or more computer programs adapted to be loaded by the processor and to implement a method as in the first aspect.
In the embodiment of the application, aiming at the determined one or more target feature sets, when the feature values under the target feature sets meet the requirements of the target wind control scene, the computer equipment generates a risk control rule based on the one or more target feature sets and corresponding feature dimensions which meet the requirements of the target wind control scene, and each relevant target feature set is obtained by screening the feature values of the feature sets collected under the target wind control scene, so that the risk control rule generated by the computer equipment is generated based on the relevant feature information under the target wind control scene. That is, the risk control rule generated by the embodiment of the application has a strong correlation with the target wind control scene, so that the risk control rule generated by the embodiment of the application can be well adapted to the application of the target wind control scene, and the accuracy of the computer equipment in the risk control processing under the target wind control scene can be ensured to a certain extent based on the risk control rule, that is, the risk control rule under the target wind control scene can be effectively generated by adopting the embodiment of the application.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic diagram of a rule generating system according to an embodiment of the present application;
FIG. 2 is a schematic flow chart of a rule generation method provided by an embodiment of the present application;
FIG. 3 is a schematic flow chart of yet another rule generation method provided by an embodiment of the present application;
FIG. 4 is a schematic diagram of a process for determining a target feature set according to an embodiment of the present application;
fig. 5 is a schematic structural diagram of a rule generating device according to an embodiment of the present application;
fig. 6 is a schematic structural diagram of a computer device according to an embodiment of the present application.
Detailed Description
The embodiment of the application provides a rule generation method which can be executed by computer equipment and is used for generating a risk control rule under a wind control scene for target business. In the method, a computer device acquires at least one feature set by acquiring feature values of a plurality of sample objects in a target wind-controlled scene, wherein one feature set is formed based on the feature values in one feature dimension. And aiming at each feature set, carrying out feature value screening processing on the feature set based on a feature value screening rule and a value range of a feature value in the corresponding feature set, thereby determining a screening feature range of the feature set and an indication parameter of a screening effect corresponding to the screening feature range, wherein the value of the indication parameter is positively correlated with the screening effect, so that the computer equipment can determine a target feature set from the feature set of which the indication parameter meets a parameter threshold value, and the number of the target feature set is at least one. Further, the computer equipment performs compliance verification on each determined target feature set under the target wind control scene, so as to obtain a verification result corresponding to each target feature set, wherein one target feature set corresponds to the verification result, and finally, the computer equipment can generate a risk control rule of the target wind control scene based on the feature dimension corresponding to the obtained target feature set and the feature value under the corresponding target feature set when a reference verification result for indicating that the feature value under the corresponding target feature set meets the requirement of the target wind control scene exists in the obtained verification result. The feature dimension adopted for generating the risk control rule and the target feature set may be one or more, and the feature value under the target feature set may be used to generate a constraint condition of the risk control rule on the feature value under the corresponding feature dimension, so that the generated risk control rule is used to indicate: and determining to perform risk control processing on the target object when the characteristic value of the target object under the characteristic dimension of the corresponding target characteristic set meets the corresponding constraint condition.
In the embodiment of the application, the risk control rule generated by the computer equipment is generated based on at least one target feature set and the feature dimension corresponding to each target feature set, and each target feature set is obtained by screening the feature values in the feature set collected under the target wind control scene, that is, each target feature set and the feature dimension corresponding to each target feature set are collected by the computer equipment in a targeted manner under the target wind control scene, so that the risk control rule generated by the computer equipment based on the target feature set and the corresponding feature dimension has stronger relevance with the target wind control scene, thereby the risk control rule generated by the embodiment of the application can be well adapted to the application of the target wind control scene, and the accuracy of the computer equipment in the risk control processing based on the risk control rule can be ensured to a certain extent, thereby effectively generating the risk control rule under the target wind control scene. In addition, as at least one target feature set adopted in the generation of the risk control rule can be adopted, and one target feature set corresponds to one feature dimension, the risk control rule is used for judging related information from at least one feature dimension, that is, the risk control rule generated by the embodiment of the application can have more comprehensive screening dimensions, the accuracy of the risk control rule in application can be further ensured to a certain extent, and the effectiveness of the risk control rule in a target wind control scene is improved.
In one embodiment, the target wind control scenario may be a risk control scenario under a target service, where the target service refers to a service with risk control requirements. Risk control refers to the process of taking various measures and methods to reduce the various possibilities of occurrence of a risk event, which may be understood as: events that can bring about a loss to the targeted business (including, but not limited to, economic loss, wind assessment loss, user volume loss, etc.). In practical applications, the risk event may be caused by a risk object, which in the embodiment of the application is determined based on a risk control rule. When determining a risk object by using a risk control rule, it is generally required to determine from one or more feature dimensions of the object, where the one or more feature dimensions are determined by a computer device based on a verification result of a compliance verification of each target feature set, where the compliance verification is used to determine whether a feature value under the target feature set meets a requirement of a target wind-controlled scenario, where the requirement of the target wind-controlled scenario may be directly specified, or may be determined by the computer device based on a feature dimension corresponding to the target feature set. In addition, in a specific implementation manner, the feature values included in the target feature set are feature values in a screening feature range corresponding to the feature set, where the feature values are in the feature set corresponding to the feature dimension, and the screening feature range corresponding to the feature set may be determined based on a feature value screening rule. And illustratively, a screening feature range may be used to indicate: the object needing to be subjected to risk control in the target wind control scene is in a value range of the characteristic value in the characteristic dimension corresponding to the characteristic set.
In one embodiment, the rule generating method provided by the embodiment of the present application may be applied to the rule generating system shown in fig. 1. As shown in fig. 1, the rule generating system may include a plurality of client running devices and a rule generating device (i.e., the above-described computer device for executing the rule generating method), and the rule generating device establishes communication connections with the respective client running devices. Each client running device is used for running a target client, the target client is a client for providing risk control processing under a target wind control scene of a target service, different sample objects can be associated with the target client running in different client running devices, so that the rule generating device can acquire characteristic values of a plurality of sample objects in at least one characteristic dimension from each client running device based on communication connection with each client running device, at least one characteristic set is constructed and obtained, and further the rule generating device can perform relevant processing based on the at least one characteristic set, and finally the generation of a risk control rule is realized.
The client running device may be a terminal device, and the rule generating device may be a terminal device or a server. In a specific implementation, the terminal device includes, but is not limited to, a smart phone, a portable computer, a tablet computer, an on-board terminal, an intelligent home appliance, and the like. When the rule generating device is a terminal device, the rule generating device can also operate the target client, so that the rule generating device can acquire the characteristic value of the sample object from the client operating device, and can acquire the characteristic value of the sample object from the rule generating device, the acquisition range of the characteristic set is expanded, and the acquisition efficiency of the characteristic set can be improved to a certain extent. The server may specifically include, but is not limited to: the cloud server comprises an independent physical server, a server cluster or a distributed system formed by a plurality of physical servers, and a cloud server for providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDNs (Content Delivery Network, content delivery networks), basic cloud computing services such as big data and artificial intelligent platforms and the like.
Referring to fig. 2, fig. 2 is a schematic flowchart of a rule generating method according to an embodiment of the present application. The rule generating method may still be performed by the above-described computer device (or rule generating device), and as shown in fig. 2, the method may include steps S201-S204:
s201, at least one feature set is obtained, one feature set corresponds to one feature dimension, and any feature set contains feature values of different sample objects in corresponding feature dimensions; the characteristic value of any sample object is acquired under the target wind control scene.
In one embodiment, the computer device may first determine at least one feature dimension to be referred to when generating the risk control rule in the target wind control scenario, and further construct a corresponding feature set by using feature values of the plurality of sample objects in each feature dimension, thereby obtaining at least one feature set. In the embodiment of the present application, the feature dimension refers to a feature of the object, such as gender, age, height, etc., which is a feature dimension, and the feature value in the feature dimension is specific information for describing the corresponding feature of the object, such as gender, where the specific information may be described by a man or a woman, and the feature value in the feature dimension of the gender may be any one of the man or the woman. In addition, the feature set corresponding to each feature dimension is determined based on the same plurality of sample objects, and the feature value of each sample object under each feature dimension can be obtained from the historical wind control data generated under the target wind control scene, so that the embodiment of the application can generate the risk control rule by performing machine learning on the historical wind control data. Wherein the historical wind control data refers to data generated before the computer device acquires the at least one feature set, and the historical wind control data at least comprises feature values of the plurality of sample objects in the at least one feature dimension.
In practical application, any sample object may be an object (called a risk control object) that performs a risk control process in a historical time, or may be an object (called a non-risk control object) that does not perform a risk control process, so that at least one risk control object and at least one non-risk control object may exist in a plurality of sample objects adopted in the embodiment of the present application. Then, the computer device may collect, for the plurality of sample objects, feature values of each feature dimension, and sum up features of the feature values of the risk control objects in the corresponding feature dimensions, so as to generate a risk control rule based on the sum up features and the corresponding feature dimensions, so as to accurately determine, based on the risk control rule, an object that needs to be subjected to risk control processing in the target wind control scene. Optionally, to facilitate distinguishing between risk control objects and non-risk control objects in the plurality of sample objects, the embodiment of the present application may generate object labels for each sample object, so as to use the object labels to indicate whether the corresponding sample object is a risk control object.
S202, screening the characteristic values in any characteristic set based on characteristic value screening rules and combining the value range of the characteristic values in any characteristic set, and performing characteristic value screening processing on the any characteristic set to obtain a screening characteristic range of any characteristic set and an indication parameter of the screening characteristic range of any characteristic set corresponding to the screening effect.
In one embodiment, the feature value screening rule may be associated with a feature dimension, so that different feature dimensions may be associated with different feature value screening rules, and thus, when the computer device performs feature value screening processing on each feature set, different feature value screening rules may be adopted, and different feature dimensions may be associated with different feature value screening rules, and thus, the feature value screening rules may be set to a greater extent with reference to the characteristics of related feature dimensions, so that pertinence of the feature value screening rules may be improved, and improvement is brought to a certain extent for the effect of the feature value screening processing. In other embodiments, the feature value filtering rules may also be associated with a type of feature value, which may include, but is not limited to, any of numeric and enumerated feature values. The numerical characteristic value refers to a characteristic value expressed by a numerical value, and the numerical characteristic value can be any value in a certain continuous numerical range. For example, the characteristic value corresponding to the characteristic dimension of the height is a numerical characteristic value, and the value of the characteristic dimension can be any value between 0 and 220. In addition, the enumerated type feature value refers to a feature value represented by discrete characters, and the characters may be composed of one or more of numerals, letters, symbols, and the like. For the enumerated feature values, the value of the enumerated feature values can be any candidate value contained in the candidate value set of the corresponding feature dimension. For example, the feature value of the sex in the feature dimension is an enumerated feature value, and the value of the feature value can be represented by a man or a woman, and the man and the woman are candidates in the candidate value set corresponding to the feature dimension.
Based on the above description, in one embodiment, the range of values of the feature values in the feature set may also be determined based on the type of the feature values in the feature set and the values of the feature values. Specifically, for a feature set in which the type of feature value is a numerical feature value, the range of values of the feature value in the feature set may be a continuous numerical range. Alternatively, when determining the value range corresponding to the feature set, a value interval formed by the feature value with the smallest value and the feature value with the largest value in the feature set may be used as the value range of the feature value in the feature set. That is, for the feature set a whose feature value type is a numerical feature value, if the feature value with the smallest value in the feature set a is a and the feature value with the largest value is b, the value range of the feature value in the feature set a is [ a, b ]. Correspondingly, for the feature set with the enumerated feature value type, the value range of the feature value in the feature set is represented by a candidate feature value set, and the candidate feature value set at least comprises one candidate feature value. Optionally, when determining the value range corresponding to the feature set, one or more feature values obtained after the feature set is de-duplicated may be used as feature values in the candidate feature value set corresponding to the feature set, and then the candidate feature values in the candidate feature value set are used as the value range of the feature set.
In one embodiment, when determining the screening feature range for the feature set formed by the numerical feature values, the computer device may perform feature value screening processing on the feature values in the feature set after determining the feature value screening rule and the value range of the feature values in the feature set, so as to determine at least one reference boundary value, and further generate the screening feature range of the feature set based on the value range of the at least one reference boundary value corresponding to the feature set. The at least one reference boundary value may include a minimum reference value of the feature values indicated by the screening feature range and/or a maximum reference value of the feature values indicated by the screening feature range. For example, for a feature set with a range of values of [ c, d ] and a determined reference boundary value of e, the screening feature range of the feature set may be one or both of [ c, e ] and [ e, d ]. It will be appreciated that in practice, the screening feature range may include one or more, and embodiments of the present application are not limited in this regard. In addition, a screening feature range may correspondingly determine an indication parameter of a screening effect, optionally, the indication parameter includes a parameter name and a parameter value, and the indication parameter name corresponding to each screening feature range is the same, and the parameter value may be different. The indication parameter of the screening feature range of any feature set may be used to indicate that the value included in the feature set is in the feature value outside the corresponding screening feature range, which is the ratio between the number of feature values of the non-risk control object and the number of feature values of the risk control object, where the ratio is positively correlated with the screening effect of the screening feature range.
S203, determining a target feature set from the feature set of which the corresponding indication parameter meets a parameter threshold according to the indication parameter of the screening effect corresponding to each feature set, and carrying out compliance verification on the feature value under the obtained target feature set under a target wind control scene to obtain at least one verification result.
In one embodiment, the indication parameter meets a parameter threshold, which means that the parameter value of the indication parameter is greater than or equal to the parameter threshold, and the parameter threshold may be a preset fixed parameter value threshold, or may be determined based on each indication parameter according to a preset rule after determining the indication parameter of the screening effect corresponding to each feature set. Illustratively, the preset rules may be: and taking the maximum parameter value in the determined indication parameters as a parameter threshold. For a feature set for which the indication parameter of the screening effect meets the parameter threshold, the manner in which the computer device determines the target feature set from the feature set may be: and taking the characteristic value of the characteristic set, of which the characteristic value is in the corresponding screening characteristic range, as the characteristic value contained in the target characteristic set, and constructing the target characteristic set.
In addition, it should be noted that, in practical application, since one or more screening feature ranges corresponding to one feature set may be provided, and in the one or more screening feature ranges, an indication parameter of a screening effect of one or more screening feature ranges may meet a parameter threshold, in this embodiment of the present application, one or more target feature sets determined for one feature set may also be provided, and one target feature set of the feature set is obtained based on one screening feature range. In addition, in at least one feature set, there may be one or more feature sets in which the indication parameter of the screening effect corresponding to the screening feature range satisfies the parameter threshold, so that the target feature set determined by the computer device may also be one or more.
In one embodiment, the purpose of the computer device to perform compliance verification on the target feature set is to determine whether the feature values under the target feature set meet the requirements of the target wind control scenario. One verification result can be obtained by performing compliance verification on one target feature set, but in the embodiment of the application, the number of target feature sets can be one or more, so that one or more (i.e., at least one) verification results can be obtained after the compliance verification is performed on the obtained target feature set, and each verification result is used for indicating whether the feature value under the corresponding target feature set meets the requirement of the target wind control scene.
S204, in the obtained at least one verification result, when the reference verification result indicates that the characteristic value under the corresponding target characteristic set meets the requirement of the target wind control scene, generating a risk control rule of the target wind control scene based on the characteristic dimension corresponding to the obtained target characteristic set and the characteristic value under the corresponding target characteristic set.
In one embodiment, the total number of risk control rules generated by the computer device may be one, while there may be at least one target feature set for generating the risk control rules, and one target feature set corresponds to one feature dimension, such that the risk control rules may be generated based on the at least one feature dimension and feature values under the respective target feature set, wherein the feature values under the target feature set may be regarded as constraints on the respective feature dimension. That is, when the embodiment of the present application generates only a single risk control rule, each feature dimension and the corresponding constraint condition may be specified under the same risk control rule. Specifically, constraint conditions of each feature dimension may be deployed in the same risk control rule in a logical and manner, where deployment in a logical and manner refers to: in the deployment stage, constraint conditions of all feature dimensions for generating the risk control rule are logically connected in series, so that when feature values of any object under all feature dimensions meet corresponding constraint conditions, the computer equipment can determine to perform risk control processing on the object. It is easy to see that the generated single risk control rule is deployed in a logical and mode, so that complex logic is not required to be considered during deployment, the workload of the risk control rule in the deployment stage can be reduced, the rule deployment efficiency is improved, and the difficulty of the rule deployment stage is reduced.
In addition, in one embodiment, when the computer device generates the risk control rule, the computing device may first determine a value range of the feature values in each target feature set, and generate constraint conditions of the feature values under the feature dimensions corresponding to the corresponding target feature sets based on the determined value range, so that the computing device may generate the risk control rule based on the feature dimensions of each target feature set and the corresponding constraint conditions. The risk control rule is used to determine: and the object with the feature value meeting the constraint condition under the corresponding feature dimension is an object needing risk control processing.
In the embodiment of the application, aiming at the determined one or more target feature sets, when the feature values under the target feature sets meet the requirements of the target wind control scene, the computer equipment generates a risk control rule based on the one or more target feature sets and corresponding feature dimensions which meet the requirements of the target wind control scene, and each relevant target feature set is obtained by screening the feature values of the feature sets collected under the target wind control scene, so that the risk control rule generated by the computer equipment is generated based on the relevant feature information under the target wind control scene. That is, the risk control rule generated by the embodiment of the application has a strong correlation with the target wind control scene, so that the risk control rule generated by the embodiment of the application can be well adapted to the application of the target wind control scene, and the accuracy of the computer equipment in the risk control processing under the target wind control scene can be ensured to a certain extent based on the risk control rule, that is, the risk control rule under the target wind control scene can be effectively generated by adopting the embodiment of the application.
Based on the rule generating method shown in fig. 2, another rule generating method is also provided in the embodiment of the present application. Referring to fig. 3, fig. 3 is a schematic flow chart of the rule generating method. The rule generating method may still be performed by the above-described computer device (or rule generating device), and as shown in fig. 3, the method may include steps S301-S306:
s301, acquiring at least one feature set, wherein one feature set corresponds to one feature dimension, any feature set comprises feature values of N different sample objects under the corresponding feature dimension, and N is an integer larger than 1; the characteristic value of any sample object is acquired under the target wind control scene.
In one embodiment, the at least one feature set collected by the computer device may include a first feature set, and the feature value in the feature dimension corresponding to the first feature set is a numerical feature value. In other embodiments, the at least one feature set may further include a second feature set, and the feature value in the feature dimension corresponding to the second feature set is an enumerated feature value. Furthermore, of the N sample objects employed by each feature set, there is at least one risk control object and at least one non-risk control object.
S302, based on a characteristic value screening rule, combining the value range of the characteristic value in any characteristic set, and performing characteristic value screening processing on any characteristic set to obtain the screening characteristic range of any characteristic set.
In one embodiment, when the any feature set is the first feature set, the computer device may determine that the corresponding sample object is a feature value of the risk control object in the first feature set, and select, from the feature values of the risk control object, a target feature value in which an arrangement sequence of the feature value values is located in a target sequence, and further generate a screening feature range of the first feature set based on a value range of the feature value in the first feature set and the target feature value. The arrangement sequence of the feature value values can be determined according to the sequence of the feature value values from large to small, or according to the sequence of the feature value values from small to large, and the target sequence for determining the target feature value can be a preset fixed value, and optionally, the fixed value can be associated with a determination mode of the arrangement sequence of the feature value values, so that the target sequence adopted when the arrangement sequence is determined according to the sequence of the feature value values from large to small is inconsistent with the target sequence adopted when the arrangement sequence is determined according to the sequence of the feature value values from small to large, and the determined target feature value has higher reference value. Of course, in other embodiments, the target order may be dynamically determined according to the number of non-repeated feature values in the first feature set, for example: and taking the numerical value corresponding to half of the number as the serial number value corresponding to the target sequence. The embodiment of the application does not limit the way in which the target order is determined.
In addition, in the embodiment of the application, the screening feature range determined by the computer equipment based on the target feature value and the value range of the first feature set is smaller than the value range of the feature value in the first feature set. That is, when generating the feature values within the screening feature range, the feature values are selected less than the feature values within the corresponding value range. In practical application, the screening feature range is smaller than the corresponding value range, which may include the following three cases: (1) The minimum characteristic value indicated by the screening characteristic range is larger than the minimum characteristic value indicated by the corresponding value range; (2) Screening the maximum characteristic value indicated by the characteristic range, wherein the maximum characteristic value is smaller than the maximum characteristic value indicated by the corresponding value range; (3) The minimum characteristic value indicated by the screening characteristic range is larger than the minimum characteristic value indicated by the corresponding value range, and the maximum characteristic value indicated by the screening characteristic range is smaller than the maximum characteristic value indicated by the corresponding value range.
Based on the above description, in one embodiment, when the computer device generates the screening feature range of the first feature set based on the value range of the feature value in the first feature set and the target feature value, the target feature value may be used as the minimum feature value indicated by the screening feature range, and at this time, the computer device may take the value of the maximum feature value indicated by the corresponding value range, so as to construct the screening feature range of the first feature set. Optionally, the computer device may also take the target feature value as a maximum feature value indicated by the screening feature range, so that the computer device may take a minimum feature value indicated by the corresponding value range as a minimum feature value indicated by the feature screening range, and further construct a screening feature range for obtaining the first feature set. That is, in the embodiment of the present application, the computer device may implement the construction of the screening feature range of the first feature set by determining the reference feature value indicated by the value range corresponding to the first feature set, and determining the target feature value from the first feature set. The reference feature value may include one or two of a minimum feature value indicated by the corresponding value range and a maximum feature value indicated by the value range, so that the computer device may construct and obtain the corresponding one or two screening feature ranges by using the value range formed by the target feature value and the reference feature value as one screening feature range of the first feature set.
The manner in which the screening feature ranges are determined for the first feature set in the embodiments of the present application is described in detail below with reference to specific examples. In this example, assume that the first feature set uses C 0 And C represents 0 ={x 1 ,x 2 ,x 3 ,…,x N X, where x i (i may be 1, 2, …, N) refers to the feature values in the first feature set, each feature value may correspond to a tag information y, where the tag information is used to indicate whether the sample object corresponding to the feature value is a risk control object, and the sample object corresponding to the feature value refers to the sample object used when the feature value is collected. Illustratively, y=1 indicates that the tag information indicates that the sample object corresponding to the respective feature value is a risk control object, and y=0 indicates that the tag information indicates that the sample object corresponding to the respective feature value is not a risk control object.
In this case, the computer device may first select from C 0 The smallest eigenvalue is determined (usedRepresented), and the maximum characteristic value (in +.>Representation) to ∈>As a means ofFirst feature set C 0 Is a range of values. Further, the computer device may be for C 0 In the first feature set obtained after the de-duplication process, determining each feature value with tag information of y=1, so as to select a second feature value from each feature value with y=1 according to the sequence of the feature values from small to large (namely, the second minimum value in each feature value with y=1 is used- >Representation) and the next to last characteristic value (i.e. the second maximum of the individual characteristic values of y=1, with +.>Representation) as target characteristic values, respectively, so that the computer device can add +.>And +.>Two screening feature ranges as the first feature set.
In one embodiment, when determining the screening feature range for the first feature set, the computer device may further generate a candidate range based on the value range of the feature values in the first feature set and the target feature value, so as to use the candidate range as the screening feature range of the first feature set when determining that the number of feature values in the candidate range in the first feature set is greater than or equal to the first preset number. And when the number of the characteristic values in the candidate range in the first characteristic set is smaller than the first preset number, the computer equipment determines a new target characteristic value in the first characteristic set so that the number of the characteristic values in the value range formed by the new target characteristic value and the reference characteristic value in the characteristic values contained in the first characteristic set is the first preset number. In this case, the computer device may then generate a filtered feature range for the first feature set based on the value range of the feature values in the first feature set and the new target feature value. Therefore, in the embodiment of the application, the value of at least the first preset number of feature values in the first feature set is located in the screening feature range, so that the computer equipment can determine a larger screening feature range under the constraint of the first preset number, the phenomenon that the screening feature range is fitted when being applied is reduced, and the screening feature range is further enabled to have higher generalization and better screening effect.
The first preset number may be determined according to a preset ratio between the number of feature values in the screening feature range in the first feature set and the number of feature values in the first feature set, if the preset ratio is p, the first preset number is a multiplication result of the number of feature values in the first feature set and the preset ratio p, and p=5% is an example. In practical applications, the number of feature values in the candidate range in the first feature set may be determined by using the manner shown in equation 1 and/or equation 2, and whether the number is greater than or equal to the first preset number.
In equation 1, # represents the number of elements in the correlation set, C 0 A first set of features is represented and,a screening feature range representing the first feature set, N representing C 0 And p represents a preset ratio.
In equation 2, # represents the number of elements in the correlation set, C 0 A first set of features is represented and,another screening feature range representing the first feature setN represents C 0 And p represents a preset ratio. It should be noted that, in the embodiment of the present application, the value of p in the formula 1 and the formula 2 is not limited, and may be adaptively changed according to the application scenario.
Further, when the number of feature values in the candidate range in the first feature set is smaller than the first preset number, the computer device may generate a new feature screening range by re-determining the target feature value so that the number of feature values corresponding to the feature screening range satisfies the first preset number. Specifically, the computer device may perform the deduplication processing on each feature value in the first feature set, then arrange the feature values after the deduplication processing in order from large to small or from small to large, and determine, based on each feature value after the arrangement, a score corresponding to a minimum feature value indicated by a value range of the first feature set, and further use, as the target feature value, a feature value whose difference between the score and the score is p (illustratively, the target feature value isRepresentation) so that the newly determined screening range of the characteristic can be used +.>And (3) representing. Similarly, the computer device may determine a score corresponding to the largest feature value indicated by the value range of the first feature set, and further determine, as the target feature value, a feature value whose difference between the score and the score is p, of the arranged feature values (the target feature value is exemplified by- >Representation) so that the newly determined screening range of the characteristic can be used +.>And (3) representing.
In one embodiment, when the any feature set is the second feature set, the computer device may perform a deduplication process on the feature values in the second feature set when determining the screening feature range, so as to obtain a deduplicated second feature set, where the feature values in the deduplicated feature set are at least one feature value that is not repeated. For each feature value in the second feature set after the duplication removal, the computer device may determine the number of risk control objects in the sample object corresponding to each feature value, so as to determine, from the second feature set after the duplication removal, feature values with the number of corresponding risk control objects being less than or equal to a second preset number, and further use a value range formed by other feature values in the second feature set after the duplication removal, except for the determined feature values, as a screening feature range of the second feature set. The second preset number may be a preset fixed value, or may be determined based on repeated feature values corresponding to the feature values in the second feature set, for example, the computer device may use the number of repetitions of the feature value with the smallest number of repetitions as the second preset number.
For ease of understanding, the manner in which the second preset number, and the range of screening features of the second feature set, are constructed is described below in connection with specific examples. In this example, assume that the second feature set uses C 1 Representation, C 1 ={D 1 、D 2 、D 3 ,…,D n And repeated eigenvalues may be included in C1. The feature set obtained by de-duplicating the second feature set is assumed to be D_C 1 And D_C 1 ={D 1 、D 2 ,D 4 ,D m Then the computer device may be based on C 1 Each characteristic value and corresponding label information in the database, and determining D_C 1 The number of risk sample objects corresponding to each feature value in the database, namely, the method comprises the following steps: the characteristic value is D 1 Is characterized by the number of risk control objects, the characteristic value is D 2 Is characterized by the number of risk control objects, the characteristic value is D 4 Is the number of risk control objects and has a characteristic value of D m Risk controlling the number of objects. Further, the computer device may take the determined minimum value of the number as a second preset value of the number, and perform the second feature after the duplication removalAnd (3) concentrating, removing the characteristic values of which the number of the corresponding risk control objects is smaller than or equal to the second preset number, so as to obtain a screening characteristic range, wherein the screening characteristic range is also a characteristic value set. For example, assume that the culled feature value is D 2 Then the screening feature range will be { D 1 ,D 4 ,D m }。
S303, determining the first number of risk control objects with the corresponding characteristic values out of the screening characteristic range of any characteristic set and determining the second number of non-risk control objects with the corresponding characteristic values out of the screening characteristic range of any characteristic set in the N sample objects.
S304, determining indication parameters of the screening feature range of any feature set corresponding to the screening effect according to the first number and the second number.
In one embodiment, if the first number is a positive integer, the computer device may use a ratio between the second number and the first number as a parameter value of the indication parameter of the screening effect corresponding to the screening feature range of any feature set. Correspondingly, if the first number is 0, the computer device may use the addition result between the second number and the preset value as the parameter value of the indication parameter of the screening effect corresponding to the screening feature range. The parameter value of the indication parameter is positively correlated with the corresponding screening effect, that is, the larger the parameter value of the indication parameter is, the better the screening effect of the corresponding screening characteristic range is. The screening effect mainly refers to the difference between the number of the characteristic values of the non-risk control objects and the number of the characteristic values of the risk control objects in the characteristic values filtered from the corresponding characteristic sets by adopting the screening characteristic range, and the larger the difference is, which means that the screening characteristic range can filter the number of more non-risk control objects while filtering fewer risk control objects, so that the target characteristic set generated based on the characteristic set can contain the characteristic values of more risk control objects, further the risk control rule generated based on the characteristic values under the target characteristic set can be used for determining more and more accurate objects needing risk control processing, and the risk control rule generated by the embodiment of the application has higher application value.
S305, determining a target feature set from the feature set of which the corresponding indication parameter meets a parameter threshold according to the indication parameter of the screening effect corresponding to each feature set, and carrying out compliance verification on the feature value under the obtained target feature set under a target wind control scene to obtain at least one verification result.
In one embodiment, the parameter threshold may be determined based on the indication parameters of the screening effect corresponding to each feature set, and specifically, the computer device may select, from the indication parameters of the screening effect corresponding to each feature set, a target indication parameter whose parameter values are arranged in the reference order, and further use the parameter value of the target indication parameter as the parameter threshold. The arrangement order of the parameter values may be determined in the order from large to small, or may be determined in the order from small to large, and when the arrangement order is determined in the order from large to small, the reference order may be the X-th bit, and when x=1, the parameter threshold is the maximum parameter value among the parameter values. Similarly, when determined in order from small to large, the reference order may be the Y-th bit, and when the value of Y is the same as the number of parameter values, the parameter threshold is also the largest parameter value among the respective parameter values. Where X and Y are both positive integers, and X may be equal to Y, for example.
In one embodiment, the number of feature sets indicating that the parameter meets the parameter threshold may be one or more, and in order to reduce the calculation amount of the computer device when generating the risk control rule, thereby reducing the difficulty of the computer device when generating the risk control rule and improving the rule generation efficiency, in the embodiment of the present application, the computer device may determine the target feature set from one feature set indicating that the parameter meets the parameter threshold each time the target feature set needs to be determined. Of course, in order to ensure the quality of the risk control rule, the computer device may also determine the target feature set from the respective feature sets indicating that the parameter satisfies the parameter threshold when the target feature set needs to be determined, which is not limited by the embodiment of the present application. For the determined target feature set, when the computer device performs compliance verification on the target feature set, the risk control object may be determined from the sample objects corresponding to the target feature set according to the object labels of the sample objects corresponding to the feature values in the target feature set, so as to generate a verification result for indicating that the feature values under the corresponding target feature set meet the requirements of the target wind control scene when the ratio between the number of the determined risk control objects and the number of the feature values in the target feature set is greater than or equal to a preset ratio (illustratively, the preset ratio is denoted by n). Optionally, when the computer device performs compliance verification on the target feature set, a verification result for indicating that the feature value under the corresponding target feature set meets the requirement of the target wind control scene may be generated when the determined number of risk control objects is smaller than a preset threshold (the preset threshold is represented by r in an exemplary manner).
S306, when the reference verification result indicates that the characteristic value under the corresponding target characteristic set meets the requirement of the target wind control scene in the obtained at least one verification result, generating a risk control rule of the target wind control scene based on the characteristic dimension corresponding to the obtained target characteristic set and the characteristic value under the corresponding target characteristic set.
In one embodiment, when the feature value under the target feature set is a numeric feature value, the computer device may generate a risk control rule based on the range of values of the feature value in the target feature set and the feature dimension corresponding to the target feature set, where the risk control rule is used to indicate: and determining that risk control processing is required to be performed on the target object when the characteristic value of the target object in the corresponding characteristic dimension is in the value range corresponding to the target characteristic set. When the feature value under the target feature set is the enumerated feature value, the computer device may generate a risk control rule based on the feature value in the target feature set and the feature dimension corresponding to the target feature set, where the risk control rule is used to indicate: and determining that risk control processing is required to be carried out on the target object when the characteristic value of the target object under the corresponding characteristic dimension is not the characteristic value under the target characteristic set. It will be appreciated that when the target feature set (assumed to be a first target feature set) including both the numeric feature values and the target feature set (assumed to be a second target feature set) including the enumerated feature values for generating the risk control rule, the risk control rule generated by the computer device based on these target feature sets is used to indicate: and determining that the risk control processing is required to be performed on the target object when the characteristic value of the target object under the characteristic dimension corresponding to the first target characteristic set is in the value range corresponding to the target characteristic set and the characteristic value of the target object under the characteristic dimension corresponding to the second target characteristic set is not the characteristic value under the target characteristic set.
In one embodiment, when the obtained at least one verification result does not have the characteristic value under the corresponding target characteristic set indicated by the reference verification result to meet the requirement of the target wind control scene, determining the target sample object corresponding to each characteristic value in the target characteristic set, further adopting the characteristic value of the target sample object to update the characteristic value respectively contained in each characteristic set in the at least one characteristic set to obtain an updated characteristic set corresponding to the corresponding characteristic set, so that the computer equipment can take the obtained updated characteristic set as the new at least one characteristic set, and adopt the new at least one characteristic set to obtain the new target characteristic set, and generate the risk control rule based on the characteristic dimension of the corresponding target characteristic set and the characteristic value under the corresponding target characteristic set when the characteristic value under the new target characteristic set meets the requirement of the target wind control scene. Therefore, in the embodiment of the application, the computer equipment needs to trigger and generate the risk control rule when the determined characteristic value under the target characteristic set meets the requirement of the target wind control scene, so that the condition of determining the target characteristic set for multiple times can exist in the embodiment of the application.
One way in which the risk control rule is generated in accordance with an embodiment of the present application is described below in connection with the determination flow of the target feature set of fig. 4. Referring to fig. 4, the computer device may first determine a target feature set based on the obtained at least one feature set, and if the determined feature value under the target feature set meets the requirement of the target wind-controlled scene, consider that the target feature set meets the termination condition, and further generate the risk control rule based on the feature dimension of the target feature set and the feature value under the target feature set. It should be noted that, if there are a plurality of target feature sets determined for the first time, and at least one target feature set exists among the plurality of target feature sets to satisfy the termination condition, the computer device may select any target feature set from the at least one target feature set to generate the risk control rule. Correspondingly, if the first determined target feature set does not meet the termination condition, updating at least one feature set acquired for the first time based on the sample object (namely, the target sample object) corresponding to each feature value in the target feature set, so that the feature values in each feature set after updating are all feature values of the target sample object in corresponding feature dimensions. Further, the computer device may redetermine the target feature set based on the updated feature sets, and it may be understood that if the target feature set determined for the second time still does not meet the termination condition, the updated feature sets are continuously updated again based on the current target feature set, and the target feature set is continuously determined based on the updated feature sets until the target feature set meets the corresponding termination condition. Assuming that the target feature set determined at the Z-th time satisfies the termination condition, the computer device may generate the risk control rule based on the feature values under each of the target feature sets determined at the 1 st to Z-th times and the corresponding feature dimensions. In practical applications, the termination condition may also be a number of cycles, where the number of cycles may be preset, and z is assumed to be represented by z, and z is a positive integer, so that if the current target feature set is determined after the z-th update of the feature set, the current determined target feature set is considered to satisfy the termination condition.
It should be noted that, if there are multiple target feature sets determined for the first time and each target feature set does not meet the termination condition, the computer device may update, for each target feature set, the obtained at least one feature set based on the target feature set, so as to determine the target feature set for the second time under the feature set that corresponds to the update. If any target feature set exists in the target feature set determined for the second time and meets the termination condition, the computer device may generate a risk control rule based on the feature values under the target feature set, the feature values under the corresponding target feature set determined for the first time, and the feature dimensions of the target feature sets. The corresponding target feature set determined for the first time refers to: when the method is used for determining the target feature set of the present time, the update basis of each feature set is used. That is, these feature sets are updated according to which target sample object corresponding to which target feature set is the corresponding target feature set.
For example, assuming that at least one of the obtained feature sets has A, B and C, and that the first determined target feature set has A1 and B1, and that neither A1 nor B1 satisfies the termination condition, the computer device will update the feature set A, B, C based on the target sample object corresponding to A1, respectively, to obtain updated feature sets A1, B1 and C1, and then redetermine the target feature set based on the feature sets A1, B1 and C1 (assuming that the determined target feature set is a 2). In addition, the computer device further updates the feature set A, B, C based on the target sample object corresponding to B1, so as to obtain updated feature sets A2, B2 and C2, and then redetermines the target feature set based on the feature sets A2, B2 and C2 (assuming that the determined target feature set is B2), then based on determining whether there is a target feature set satisfying the termination condition in A2 and C2, if it is determined that A2 satisfies the termination condition, the computer device generates a risk control rule based on the feature value under A2, the feature value under a1 and the feature dimension corresponding to a1 and A2. Similarly, if b2 is determined to meet the termination condition, the computer device generates a risk control rule based on the feature value under b2, the feature value under b1, and the feature dimensions corresponding to b1 and b 2.
In the embodiment of the application, aiming at the determined one or more target feature sets, when the feature values under the target feature sets meet the requirements of the target wind control scene, the computer equipment generates a risk control rule based on the one or more target feature sets and corresponding feature dimensions which meet the requirements of the target wind control scene, and each relevant target feature set is obtained by screening the feature values of the feature sets collected under the target wind control scene, so that the risk control rule generated by the computer equipment is generated based on the relevant feature information under the target wind control scene. That is, the risk control rule generated by the embodiment of the application has a strong correlation with the target wind control scene, so that the risk control rule generated by the embodiment of the application can be well adapted to the application of the target wind control scene, and the accuracy of the computer equipment in the risk control processing under the target wind control scene can be ensured to a certain extent based on the risk control rule, that is, the risk control rule under the target wind control scene can be effectively generated by adopting the embodiment of the application. In addition, as at least one target feature set adopted in the generation of the risk control rule can be adopted, and one target feature set corresponds to one feature dimension, the risk control rule is used for judging related information from at least one feature dimension, that is, the risk control rule generated by the embodiment of the application can have a relatively comprehensive screening dimension, the accuracy of the risk control rule in application can be ensured to a certain extent, and the risk control rule can be further effectively applied to a target wind control scene.
Based on the above related embodiments of the rule generating method of fig. 2 and fig. 3, the embodiment of the present application further proposes a rule generating apparatus, which may be a computer program (including program code) running in the above mentioned computer device (or rule generating device). In a specific embodiment, the rule generating means may be adapted to perform the rule generating method shown in fig. 2 and 3. Referring to fig. 5, the rule generating apparatus at least includes an acquiring unit 501, a feature set processing unit 502, a checking unit 503, and a generating unit 504. Wherein:
an obtaining unit 501, configured to obtain at least one feature set, where one feature set corresponds to one feature dimension, and any feature set includes feature values of different sample objects in corresponding feature dimensions; the characteristic value of any sample object is acquired under a target wind control scene;
the feature set processing unit 502 is configured to perform feature value screening processing in any feature set based on a feature value screening rule and in combination with a value range of a feature value in any feature set, to obtain a screening feature range of the any feature set, and an indication parameter of a screening effect corresponding to the screening feature range of the any feature set;
A verification unit 503, configured to determine a target feature set from feature sets corresponding to the indication parameters of the screening effect and for performing compliance verification on feature values under the obtained target feature set in the target wind control scenario, where the corresponding indication parameters meet a parameter threshold, so as to obtain at least one verification result;
and the generating unit 504 is configured to generate, when the obtained at least one verification result indicates that the feature value under the corresponding target feature set meets the requirement of the target wind-controlled scene, a risk control rule of the target wind-controlled scene based on the feature dimension corresponding to the obtained target feature set and the feature value under the corresponding target feature set.
In one embodiment, the verification unit 503 may be further specifically configured to perform:
in at least one obtained verification result, when no reference verification result indicates that the characteristic values under the target characteristic set meet the requirements of the target wind control scene, determining target sample objects corresponding to the characteristic values in the target characteristic set respectively;
adopting the characteristic values of the target sample object to update the characteristic values contained in each characteristic set in the at least one characteristic set respectively to obtain updated characteristic sets corresponding to the corresponding characteristic sets;
And taking the obtained updated feature set as a new at least one feature set, and adopting the new at least one feature set to obtain a new target feature set so that the feature value under the new target feature set meets the requirement of the target wind control scene.
In yet another embodiment, each feature set includes feature values for N sample objects in a respective feature dimension, N being an integer greater than 1; the at least one feature set includes a first feature set, and if the feature value under the feature dimension corresponding to the first feature set is a numerical feature value, the feature set processing unit 502 may be specifically configured to perform:
determining a characteristic value of which a corresponding sample object is a risk control object in the first characteristic set, and selecting a target characteristic value of which the arrangement sequence of the corresponding characteristic value values is positioned in a target sequence from the characteristic values of the risk control object;
and generating a screening feature range of the first feature set based on the value range of the feature values in the first feature set and the target feature value, wherein the screening feature range of the first feature set is smaller than the value range of the feature values in the first feature set.
In yet another embodiment, the feature set processing unit 502 may be further configured to perform:
generating a candidate range based on the value range of the characteristic values in the first characteristic set and the target characteristic value;
and determining the number of the characteristic values in the first characteristic set in the candidate range, and taking the candidate range as a screening characteristic range of the first characteristic set when the determined number is greater than or equal to a first preset number.
In yet another embodiment, the feature set processing unit 502 may be further configured to perform:
when the determined number is smaller than the first preset number, determining a new target feature value in the first feature set; the number of the characteristic values in the value range formed by the new target characteristic value and the reference characteristic value in the characteristic values contained in the first characteristic set is the first preset number;
and generating a screening feature range of the first feature set based on the value range of the feature values in the first feature set and the new target feature value.
In yet another embodiment, the feature set processing unit 502 may be further configured to perform:
determining a reference characteristic value indicated by the value range; wherein the reference characteristic value comprises one or two of a minimum characteristic value indicated by the value range and a maximum characteristic value indicated by the value range;
And taking a value range formed by the target characteristic value and the reference characteristic value as a screening characteristic range of the first characteristic set.
In yet another embodiment, each feature set includes feature values for N sample objects in a respective feature dimension, N being an integer greater than 1; the at least one feature set includes a second feature set, and if the feature value under the feature dimension corresponding to the second feature set is an enumerated feature value, the feature set processing unit 502 may be further configured to execute:
performing de-duplication treatment on the characteristic values in the second characteristic set to obtain a de-duplicated second characteristic set;
determining the number of risk control objects in the sample object corresponding to each characteristic value according to each characteristic value in the second characteristic set after the duplication removal;
determining that the number of corresponding risk control objects is less than or equal to a second preset number of feature values from the second feature set after the duplication removal;
and taking a value range formed by the other characteristic values except the determined characteristic value in the second characteristic set after the duplication removal as a screening characteristic range of the second characteristic set.
In yet another embodiment, the verification unit 503 may be specifically configured to perform:
selecting target indication parameters of which the arrangement sequence of parameter values of the indication parameters is positioned in a reference sequence from the indication parameters of the screening effect corresponding to each feature set;
and taking the parameter value of the target indication parameter as the parameter threshold value.
In yet another embodiment, the feature set processing unit 502 may be specifically configured to perform:
each feature set comprises feature values of N sample objects under corresponding feature dimensions, wherein N is an integer greater than 1; the determination mode of the indication parameter of the screening characteristic range corresponding to the screening effect of any characteristic set comprises the following steps:
determining a first number of risk control objects with corresponding characteristic values outside the screening characteristic range of any characteristic set and a second number of non-risk control objects with corresponding characteristic values outside the screening characteristic range of any characteristic set in the N sample objects;
and determining indication parameters of the screening feature range of any feature set corresponding to the screening effect according to the first quantity and the second quantity.
In yet another embodiment, the feature set processing unit 502 may be specifically configured to perform:
If the first quantity is a positive integer, taking the ratio between the second quantity and the first quantity as a parameter value of an indication parameter of the screening effect corresponding to the screening characteristic range of any characteristic set; wherein, the parameter value of the indication parameter is positively correlated with the corresponding screening effect.
In yet another embodiment, one sample object corresponds to an object tag indicating whether the corresponding sample object is a risk control object; the verification unit 503 may be further configured to specifically perform:
determining a risk control object from the sample objects corresponding to the target feature set according to the object labels of the sample objects corresponding to the feature values in the target feature set;
and when the determined number of the risk control objects is smaller than a preset threshold value, generating a verification result for indicating that the characteristic value under the corresponding target characteristic set meets the requirement of the target wind control scene.
In yet another embodiment, one sample object corresponds to an object tag indicating whether the corresponding sample object is a risk control object; the verification unit 503 may be further configured to specifically perform:
determining a risk control object from the sample objects corresponding to the target feature set according to the object labels of the sample objects corresponding to the feature values in the target feature set;
And generating a verification result for indicating that the characteristic value under the corresponding target characteristic set meets the requirement of the target wind control scene when the ratio between the determined number of the risk control objects and the number of the characteristic values in the target characteristic set is larger than or equal to a preset ratio.
According to an embodiment of the present application, each unit in the rule generating apparatus shown in fig. 5 is divided based on a logic function, and each unit may be formed by combining one or several additional units separately or all of the units, or some (some) of the units may be formed by splitting a plurality of units with smaller functions, which may achieve the same operation without affecting the implementation of the technical effects of the embodiment of the present application. In other embodiments of the present application, the rule generating apparatus may also include other units, and in practical applications, these functions may also be implemented with assistance of other units, and may be implemented with assistance of multiple units.
According to another embodiment of the present application, the rule generating apparatus shown in fig. 5 may be constructed by running a computer program (including program code) capable of executing the steps involved in the methods shown in fig. 2 and 3 on a general-purpose communication device such as a rule generating device including a processing element such as a Central Processing Unit (CPU), a random access storage medium (RAM), a read only storage medium (ROM), and the like, and a storage element, and the rule generating method of the embodiment of the present application may be implemented. The computer program may be recorded on, for example, a computer storage medium, and loaded into and run in the above-described computer apparatus through the computer storage medium.
In the embodiment of the present application, for the determined one or more target feature sets, when the feature values under the target feature sets meet the requirements of the target wind-control scenario, the generating unit 504 generates a risk control rule based on one or more target feature sets and corresponding feature dimensions related to the target wind-control scenario, where each related target feature set is obtained by screening the feature values of the feature sets collected under the target wind-control scenario by the acquiring unit 501, so that the risk control rule generated by the generating unit 504 is generated based on related feature information under the target wind-control scenario. That is, the risk control rule generated by the embodiment of the application has a strong correlation with the target wind control scene, so that the risk control rule generated by the embodiment of the application can be well adapted to the application of the target wind control scene, thereby ensuring the accuracy of the risk control processing under the target wind control scene based on the risk control rule, that is, the risk control rule under the target wind control scene can be effectively generated by adopting the embodiment of the application.
Based on the above description of the method embodiment and the apparatus embodiment, the embodiment of the present application further provides a computer device, which may also be the rule generating device mentioned in the above method embodiment, please refer to fig. 6. The computer device includes at least a processor 601 and a computer storage medium 602, with the processor 601 and computer storage medium 602 being connected by a bus or other means. Among them, the above-mentioned computer storage medium 602 is a memory device in a computer device for storing programs and data. It is understood that the computer storage media 602 herein can include both built-in storage media in a computer device and extended storage media supported by the computer device. The computer storage media 602 provides storage space that stores the operating system of the computer device. Also stored in this memory space are one or more computer programs, which may be one or more program codes, adapted to be loaded and executed by the processor 601. The computer 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; optionally, at least one storage medium located remotely from the processor. The processor 601, or CPU (Central Processing Unit ), is a computing core as well as a control core of a computer device, adapted to implement one or more computer programs, in particular adapted to load and execute one or more computer programs for implementing the respective method flows or respective functions.
In one embodiment, one or more computer programs stored in the computer storage medium 602 may be loaded and executed by the processor 601 to implement the corresponding method steps in the method embodiments described above with respect to fig. 2 and 3. In a specific implementation, one or more computer programs in the computer storage medium 602 may be loaded by the processor 601 and perform the steps of:
acquiring at least one feature set, wherein one feature set corresponds to one feature dimension, and any feature set contains feature values of different sample objects under the corresponding feature dimension; the characteristic value of any sample object is acquired under a target wind control scene;
based on a characteristic value screening rule, combining a value range of a characteristic value in any characteristic set, and performing characteristic value screening processing on the any characteristic set to obtain a screening characteristic range of the any characteristic set and an indication parameter of a screening effect corresponding to the screening characteristic range of the any characteristic set;
determining a target feature set from feature sets, corresponding to the indication parameters of the screening effect, of which the corresponding indication parameters meet parameter thresholds, and carrying out compliance verification on feature values under the obtained target feature set in the target wind control scene to obtain at least one verification result;
And when the obtained at least one verification result indicates that the characteristic value under the corresponding target characteristic set meets the requirement of the target wind control scene by the reference verification result, generating a risk control rule of the target wind control scene based on the characteristic dimension corresponding to the obtained target characteristic set and the characteristic value under the corresponding target characteristic set.
In one embodiment, the processor 601 may be further specifically configured to load and execute:
in at least one obtained verification result, when no reference verification result indicates that the characteristic values under the target characteristic set meet the requirements of the target wind control scene, determining target sample objects corresponding to the characteristic values in the target characteristic set respectively;
adopting the characteristic values of the target sample object to update the characteristic values contained in each characteristic set in the at least one characteristic set respectively to obtain updated characteristic sets corresponding to the corresponding characteristic sets;
and taking the obtained updated feature set as a new at least one feature set, and adopting the new at least one feature set to obtain a new target feature set so that the feature value under the new target feature set meets the requirement of the target wind control scene.
In yet another embodiment, each feature set includes feature values for N sample objects in a respective feature dimension, N being an integer greater than 1; the at least one feature set includes a first feature set, and if the feature value under the feature dimension corresponding to the first feature set is a numerical feature value, the processor 601 may be further specifically configured to load and execute:
determining a characteristic value of which a corresponding sample object is a risk control object in the first characteristic set, and selecting a target characteristic value of which the arrangement sequence of the corresponding characteristic value values is positioned in a target sequence from the characteristic values of the risk control object;
and generating a screening feature range of the first feature set based on the value range of the feature values in the first feature set and the target feature value, wherein the screening feature range of the first feature set is smaller than the value range of the feature values in the first feature set.
In yet another embodiment, the processor 601 may be further specifically configured to load and execute:
generating a candidate range based on the value range of the characteristic values in the first characteristic set and the target characteristic value;
and determining the number of the characteristic values in the first characteristic set in the candidate range, and taking the candidate range as a screening characteristic range of the first characteristic set when the determined number is greater than or equal to a first preset number.
In yet another embodiment, the processor 601 may be further specifically configured to load and execute:
when the determined number is smaller than the first preset number, determining a new target feature value in the first feature set; the number of the characteristic values in the value range formed by the new target characteristic value and the reference characteristic value in the characteristic values contained in the first characteristic set is the first preset number;
and generating a screening feature range of the first feature set based on the value range of the feature values in the first feature set and the new target feature value.
In yet another embodiment, the processor 601 may be further specifically configured to load and execute:
determining a reference characteristic value indicated by the value range; wherein the reference characteristic value comprises one or two of a minimum characteristic value indicated by the value range and a maximum characteristic value indicated by the value range;
and taking a value range formed by the target characteristic value and the reference characteristic value as a screening characteristic range of the first characteristic set.
In yet another embodiment, each feature set includes feature values for N sample objects in a respective feature dimension, N being an integer greater than 1; the at least one feature set includes a second feature set, and if the feature value in the feature dimension corresponding to the second feature set is an enumerated feature value, the processor 601 may be further specifically configured to load and execute:
Performing de-duplication treatment on the characteristic values in the second characteristic set to obtain a de-duplicated second characteristic set;
determining the number of risk control objects in the sample object corresponding to each characteristic value according to each characteristic value in the second characteristic set after the duplication removal;
determining that the number of corresponding risk control objects is less than or equal to a second preset number of feature values from the second feature set after the duplication removal;
and taking a value range formed by the other characteristic values except the determined characteristic value in the second characteristic set after the duplication removal as a screening characteristic range of the second characteristic set.
In yet another embodiment, the processor 601 may be further specifically configured to load and execute:
selecting target indication parameters of which the arrangement sequence of parameter values of the indication parameters is positioned in a reference sequence from the indication parameters of the screening effect corresponding to each feature set;
and taking the parameter value of the target indication parameter as the parameter threshold value.
In yet another embodiment, the processor 601 may be further specifically configured to load and execute:
each feature set comprises feature values of N sample objects under corresponding feature dimensions, wherein N is an integer greater than 1; the determination mode of the indication parameter of the screening characteristic range corresponding to the screening effect of any characteristic set comprises the following steps:
Determining a first number of risk control objects with corresponding characteristic values outside the screening characteristic range of any characteristic set and a second number of non-risk control objects with corresponding characteristic values outside the screening characteristic range of any characteristic set in the N sample objects;
and determining indication parameters of the screening feature range of any feature set corresponding to the screening effect according to the first quantity and the second quantity.
In yet another embodiment, the processor 601 may be further specifically configured to load and execute:
if the first quantity is a positive integer, taking the ratio between the second quantity and the first quantity as a parameter value of an indication parameter of the screening effect corresponding to the screening characteristic range of any characteristic set; wherein, the parameter value of the indication parameter is positively correlated with the corresponding screening effect.
In yet another embodiment, one sample object corresponds to an object tag indicating whether the corresponding sample object is a risk control object; the processor 601 may be further specifically configured to load and execute:
determining a risk control object from the sample objects corresponding to the target feature set according to the object labels of the sample objects corresponding to the feature values in the target feature set;
And when the determined number of the risk control objects is smaller than a preset threshold value, generating a verification result for indicating that the characteristic value under the corresponding target characteristic set meets the requirement of the target wind control scene.
In yet another embodiment, one sample object corresponds to an object tag indicating whether the corresponding sample object is a risk control object; the processor 601 may be further specifically configured to load and execute:
determining a risk control object from the sample objects corresponding to the target feature set according to the object labels of the sample objects corresponding to the feature values in the target feature set;
and generating a verification result for indicating that the characteristic value under the corresponding target characteristic set meets the requirement of the target wind control scene when the ratio between the determined number of the risk control objects and the number of the characteristic values in the target characteristic set is larger than or equal to a preset ratio.
In the embodiment of the application, aiming at the determined one or more target feature sets, when the feature values under the target feature sets meet the requirements of the target wind control scene, the computer equipment generates a risk control rule based on the one or more target feature sets and corresponding feature dimensions which meet the requirements of the target wind control scene, and each relevant target feature set is obtained by screening the feature values of the feature sets collected under the target wind control scene, so that the risk control rule generated by the computer equipment is generated based on the relevant feature information under the target wind control scene. That is, the risk control rule generated by the embodiment of the application has a strong correlation with the target wind control scene, so that the risk control rule generated by the embodiment of the application can be well adapted to the application of the target wind control scene, and the accuracy of the computer equipment in the risk control processing under the target wind control scene can be ensured to a certain extent based on the risk control rule, that is, the risk control rule under the target wind control scene can be effectively generated by adopting the embodiment of the application.
The application also provides a storage medium, in which one or more computer programs corresponding to the rule generating method are stored, and when the processor loads and executes the one or more computer programs, the description of the rule generating method in the embodiment can be realized, which is not repeated here. The description of the advantageous effects of the same method is not repeated here. It will be appreciated that a computer program may be deployed to be executed on one or more devices that are capable of communication with one another.
It should be noted that, according to an aspect of the embodiments of the present application, there is also provided a program product or a computer program, the program product including a computer program, the computer program being stored in a computer storage medium. A processor in a computer device reads the computer program from a computer storage medium and then executes the computer program, thereby enabling the computer device to perform the methods provided in the various alternatives of the rule generation method embodiments aspects shown in fig. 2 and 3 described above.
Those skilled in the art will appreciate that implementing all or part of the above-described embodiment methods may be accomplished by way of a computer program, which may be stored in a computer storage medium and which, when executed, may comprise the steps of embodiments of the rule generating method as described above. The computer storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a random-access Memory (Random Access Memory, RAM), or the like.
It is to be understood that the foregoing disclosure is only a partial embodiment of the present application, and it is not to be limited thereto, but it is to be understood that all or a partial process for implementing the above embodiment may be modified by one skilled in the art to fall within the scope of the present application as defined by the appended claims. It should be further specifically noted that, when the embodiments of the present application are applied to specific products or technologies, if the actions of acquiring the data (such as the feature values) related to the target object or the sample object are involved, the related products or technologies need to obtain the permission or consent of the corresponding objects, and the data acquired by the related products or technologies need to meet the laws and regulations and standards of the relevant countries and regions.
Claims (15)
1. A rule generation method, comprising:
acquiring at least one feature set, wherein one feature set corresponds to one feature dimension, and any feature set contains feature values of different sample objects under the corresponding feature dimension; the characteristic value of any sample object is acquired under a target wind control scene;
based on a characteristic value screening rule, combining a value range of a characteristic value in any characteristic set, and performing characteristic value screening processing on the any characteristic set to obtain a screening characteristic range of the any characteristic set and an indication parameter of a screening effect corresponding to the screening characteristic range of the any characteristic set;
Determining a target feature set from feature sets, corresponding to the indication parameters of the screening effect, of which the corresponding indication parameters meet parameter thresholds, and carrying out compliance verification on feature values under the obtained target feature set in the target wind control scene to obtain at least one verification result;
and when the obtained at least one verification result indicates that the characteristic value under the corresponding target characteristic set meets the requirement of the target wind control scene by the reference verification result, generating a risk control rule of the target wind control scene based on the characteristic dimension corresponding to the obtained target characteristic set and the characteristic value under the corresponding target characteristic set.
2. The method according to claim 1, wherein the method further comprises:
in at least one obtained verification result, when no reference verification result indicates that the characteristic values under the target characteristic set meet the requirements of the target wind control scene, determining target sample objects corresponding to the characteristic values in the target characteristic set respectively;
adopting the characteristic values of the target sample object to update the characteristic values contained in each characteristic set in the at least one characteristic set respectively to obtain updated characteristic sets corresponding to the corresponding characteristic sets;
And taking the obtained updated feature set as a new at least one feature set, and adopting the new at least one feature set to obtain a new target feature set so that the feature value under the new target feature set meets the requirement of the target wind control scene.
3. The method of claim 1, wherein each feature set comprises feature values for N sample objects in a respective feature dimension, N being an integer greater than 1; the at least one feature set includes a first feature set, and if the feature value under the feature dimension corresponding to the first feature set is a numerical feature value, the method for obtaining the screening feature range of the first feature set includes:
determining a characteristic value of which a corresponding sample object is a risk control object in the first characteristic set, and selecting a target characteristic value of which the arrangement sequence of the corresponding characteristic value values is positioned in a target sequence from the characteristic values of the risk control object;
and generating a screening feature range of the first feature set based on the value range of the feature values in the first feature set and the target feature value, wherein the screening feature range of the first feature set is smaller than the value range of the feature values in the first feature set.
4. The method of claim 3, wherein generating the screening feature range of the first feature set based on the value range of the feature values in the first feature set and the target feature value comprises:
generating a candidate range based on the value range of the characteristic values in the first characteristic set and the target characteristic value;
and determining the number of the characteristic values in the first characteristic set in the candidate range, and taking the candidate range as a screening characteristic range of the first characteristic set when the determined number is greater than or equal to a first preset number.
5. The method according to claim 4, wherein the method further comprises:
when the determined number is smaller than the first preset number, determining a new target feature value in the first feature set; the number of the characteristic values in the value range formed by the new target characteristic value and the reference characteristic value in the characteristic values contained in the first characteristic set is the first preset number;
and generating a screening feature range of the first feature set based on the value range of the feature values in the first feature set and the new target feature value.
6. The method of claim 3, wherein generating the screening feature range of the first feature set based on the value range of the feature values in the first feature set and the target feature value comprises:
determining a reference characteristic value indicated by the value range; wherein the reference characteristic value comprises one or two of a minimum characteristic value indicated by the value range and a maximum characteristic value indicated by the value range;
and taking a value range formed by the target characteristic value and the reference characteristic value as a screening characteristic range of the first characteristic set.
7. The method of claim 1, wherein each feature set comprises feature values for N sample objects in a respective feature dimension, N being an integer greater than 1; the at least one feature set comprises a second feature set, and if the feature value under the feature dimension corresponding to the second feature set is an enumeration type feature value; the manner of deriving the screening feature range of the second feature set comprises:
performing de-duplication treatment on the characteristic values in the second characteristic set to obtain a de-duplicated second characteristic set;
determining the number of risk control objects in the sample object corresponding to each characteristic value according to each characteristic value in the second characteristic set after the duplication removal;
Determining that the number of corresponding risk control objects is less than or equal to a second preset number of feature values from the second feature set after the duplication removal;
and taking a value range formed by the other characteristic values except the determined characteristic value in the second characteristic set after the duplication removal as a screening characteristic range of the second characteristic set.
8. The method according to claim 1, wherein the manner of determining the parameter threshold comprises:
selecting target indication parameters of which the arrangement sequence of parameter values of the indication parameters is positioned in a reference sequence from the indication parameters of the screening effect corresponding to each feature set;
and taking the parameter value of the target indication parameter as the parameter threshold value.
9. The method of claim 1, wherein each feature set comprises feature values for N sample objects in a respective feature dimension, N being an integer greater than 1; the determination mode of the indication parameter of the screening characteristic range corresponding to the screening effect of any characteristic set comprises the following steps:
determining a first number of risk control objects with corresponding characteristic values outside the screening characteristic range of any characteristic set and a second number of non-risk control objects with corresponding characteristic values outside the screening characteristic range of any characteristic set in the N sample objects;
And determining indication parameters of the screening feature range of any feature set corresponding to the screening effect according to the first quantity and the second quantity.
10. The method of claim 9, wherein determining, from the first number and the second number, an indication parameter of a screening feature range of the any feature set corresponding to a screening effect comprises:
if the first quantity is a positive integer, taking the ratio between the second quantity and the first quantity as a parameter value of an indication parameter of the screening effect corresponding to the screening characteristic range of any characteristic set; wherein, the parameter value of the indication parameter is positively correlated with the corresponding screening effect.
11. The method of claim 1, wherein one sample object corresponds to an object tag indicating whether the corresponding sample object is a risk control object; and under the target wind control scene, performing compliance verification on the feature values under the obtained target feature set to obtain at least one verification result, wherein the verification result comprises the following steps:
determining a risk control object from the sample objects corresponding to the target feature set according to the object labels of the sample objects corresponding to the feature values in the target feature set;
And when the determined number of the risk control objects is smaller than a preset threshold value, generating a verification result for indicating that the characteristic value under the corresponding target characteristic set meets the requirement of the target wind control scene.
12. The method of claim 1, wherein one sample object corresponds to an object tag indicating whether the corresponding sample object is a risk control object; and under the target wind control scene, performing compliance verification on the feature values under the obtained target feature set to obtain at least one verification result, wherein the verification result comprises the following steps:
determining a risk control object from the sample objects corresponding to the target feature set according to the object labels of the sample objects corresponding to the feature values in the target feature set;
and generating a verification result for indicating that the characteristic value under the corresponding target characteristic set meets the requirement of the target wind control scene when the ratio between the determined number of the risk control objects and the number of the characteristic values in the target characteristic set is larger than or equal to a preset ratio.
13. A rule generating apparatus, comprising:
the device comprises an acquisition unit, a storage unit and a processing unit, wherein the acquisition unit is used for acquiring at least one feature set, one feature set corresponds to one feature dimension, and any feature set contains feature values of different sample objects under corresponding feature dimensions; the characteristic value of any sample object is acquired under a target wind control scene;
The feature set processing unit is used for carrying out feature value screening processing on any feature set based on a feature value screening rule and combining a value range of a feature value in the any feature set to obtain a screening feature range of the any feature set and an indication parameter of a screening effect corresponding to the screening feature range of the any feature set;
the verification unit is used for determining a target feature set from the feature set of which the corresponding indication parameter meets a parameter threshold according to the indication parameter of the screening effect corresponding to each feature set, and carrying out compliance verification on the feature value under the obtained target feature set under the target wind control scene to obtain at least one verification result;
the generation unit is used for generating a risk control rule of the target wind control scene based on the feature dimension corresponding to the obtained target feature set and the feature value under the corresponding target feature set when the reference check result indicates that the feature value under the corresponding target feature set meets the requirement of the target wind control scene.
14. A computer device, comprising:
a processor for implementing one or more computer programs;
A computer storage medium storing one or more computer programs adapted to be loaded by the processor and to implement the method of any of claims 1-12.
15. A storage medium storing one or more computer programs adapted to be loaded by a processor and to implement the method of any of claims 1-12.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211681274.2A CN116976661A (en) | 2022-12-22 | 2022-12-22 | Rule generation method, device, computer equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211681274.2A CN116976661A (en) | 2022-12-22 | 2022-12-22 | Rule generation method, device, computer equipment and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN116976661A true CN116976661A (en) | 2023-10-31 |
Family
ID=88483783
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211681274.2A Pending CN116976661A (en) | 2022-12-22 | 2022-12-22 | Rule generation method, device, computer equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116976661A (en) |
-
2022
- 2022-12-22 CN CN202211681274.2A patent/CN116976661A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109242013B (en) | Data labeling method and device, electronic equipment and storage medium | |
CN110689084B (en) | Abnormal user identification method and device | |
CN109918678B (en) | Method and device for identifying field meaning | |
CN111090807A (en) | Knowledge graph-based user identification method and device | |
CN105335375B (en) | Topics Crawling method and apparatus | |
CN112800197A (en) | Method and device for determining target fault information | |
CN106933916B (en) | JSON character string processing method and device | |
CN112579462A (en) | Test case acquisition method, system, equipment and computer readable storage medium | |
CN107451249B (en) | Event development trend prediction method and device | |
CN108076032B (en) | Abnormal behavior user identification method and device | |
CN108399266B (en) | Data extraction method and device, electronic equipment and computer readable storage medium | |
CN114048136A (en) | Test type determination method, device, server, medium and product | |
CN111126965B (en) | Auditing rule optimization method, auditing rule optimization device, computer equipment and storage medium | |
CN116432069A (en) | Information processing method, service providing method, device, equipment and medium | |
CN116976661A (en) | Rule generation method, device, computer equipment and storage medium | |
CN110544166A (en) | Sample generation method, device and storage medium | |
CN112765236B (en) | Adaptive abnormal equipment mining method, storage medium, equipment and system | |
CN115982634A (en) | Application program classification method and device, electronic equipment and computer program product | |
CN110032445B (en) | Big data aggregation calculation method and device | |
CN113342647A (en) | Test data generation method and device | |
CN113946717A (en) | Sub-map index feature obtaining method, device, equipment and storage medium | |
CN114021642A (en) | Data processing method and device, electronic equipment and storage medium | |
CN109684442B (en) | Text retrieval method, device, equipment and program product | |
CN113742116A (en) | Abnormity positioning method, abnormity positioning device, abnormity positioning equipment and storage medium | |
CN113468186A (en) | Data table primary key association method and device, computer equipment and readable storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication |