CN109886554B - Illegal behavior discrimination method, device, computer equipment and storage medium - Google Patents
Illegal behavior discrimination method, device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN109886554B CN109886554B CN201910068622.0A CN201910068622A CN109886554B CN 109886554 B CN109886554 B CN 109886554B CN 201910068622 A CN201910068622 A CN 201910068622A CN 109886554 B CN109886554 B CN 109886554B
- Authority
- CN
- China
- Prior art keywords
- model
- risk
- factor
- evaluated
- violation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000012850 discrimination method Methods 0.000 title description 2
- 230000006399 behavior Effects 0.000 claims abstract description 58
- 238000000034 method Methods 0.000 claims abstract description 35
- 238000012545 processing Methods 0.000 claims description 29
- 238000004590 computer program Methods 0.000 claims description 22
- 238000012549 training Methods 0.000 claims description 18
- 238000004422 calculation algorithm Methods 0.000 claims description 12
- 238000011156 evaluation Methods 0.000 claims description 11
- 238000010586 diagram Methods 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 9
- 238000010276 construction Methods 0.000 claims description 5
- 230000003542 behavioural effect Effects 0.000 claims description 4
- 230000002996 emotional effect Effects 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 238000012502 risk assessment Methods 0.000 abstract description 11
- 238000013473 artificial intelligence Methods 0.000 abstract 1
- 238000012216 screening Methods 0.000 description 4
- 238000010801 machine learning Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000014509 gene expression Effects 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 238000010219 correlation analysis Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000000556 factor analysis Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a method, a device, computer equipment and a storage medium for distinguishing illegal behaviors, and relates to artificial intelligence-based illegal risk prediction. Comprising the following steps: acquiring characteristic information of a main body to be evaluated; inputting the characteristic information into a violation risk scoring model to obtain a violation risk score; if the illegal risk score is larger than a set threshold value, acquiring a behavior record of the main body to be evaluated in a set time window; and inputting the behavior record into a rule model to obtain an offence label. The method has the advantage that the illegal risk assessment efficiency is higher.
Description
Technical Field
The present invention relates to the field of computer technologies, and in particular, to a method and apparatus for determining illegal behaviors, a computer device, and a storage medium.
Background
With the advent of the cloud age, big data (Big data) has attracted more and more attention. With the increasing maturity of big data processing technology, more and more application branches are generated, such as illegal risk prediction based on big data analysis.
The traditional illegal risk assessment method based on big data comprises the following steps: and locating the illegal risk information or the illegal behaviors from the mass data through data identification and logic analysis. The risk prediction method is very low in risk prediction efficiency when predicting complex data and the data quantity to be predicted is large.
Disclosure of Invention
In view of the above, it is necessary to provide a method, an apparatus, a computer device, and a storage medium for determining an offence, which can improve the offence risk assessment efficiency.
A method of offence discrimination, comprising:
Acquiring characteristic information of a main body to be evaluated;
Inputting the characteristic information into a violation risk scoring model to obtain a violation risk score;
If the illegal risk score is larger than a set threshold value, acquiring a behavior record of the main body to be evaluated in a set time window;
And inputting the behavior record into a rule model to obtain an offence label.
In one embodiment, after the entering the behavior record into the rule model to obtain the offence tag, the method further includes:
And obtaining the risk pointing key corresponding to the evaluation subject according to the association relation between the predefined illegal behavior label and the risk pointing key.
In one embodiment, the method further comprises:
Acquiring a factor tree corresponding to a main body to be evaluated, wherein nodes of the factor tree comprise factors describing characteristics of the main body to be evaluated;
Selecting factors from the factor tree as characteristic variables of the model according to target variables of the violation risk scoring model to be constructed;
And constructing a violation risk scoring model according to the determined characteristic variable and the target variable.
In one embodiment, the factor tree is marked with data positioning information and data processing information of corresponding associated data for each factor;
the constructing a violation risk scoring model according to the determined characteristic variable and the target variable comprises the following steps:
Acquiring the data positioning information and the data processing information corresponding to the characteristic variable from the factor tree;
acquiring associated data corresponding to the characteristic variable according to the data positioning information;
Preprocessing the acquired associated data according to the data processing information to obtain a training sample;
and inputting the training sample into a preselected model algorithm for supervised learning to obtain a violation risk scoring model.
In one embodiment, the selecting the factor from the factor tree according to the target variable of the violation risk scoring model to be constructed as the characteristic variable of the model includes:
calculating the discrimination capability value of each factor in the factor tree for the target variable of the violation risk scoring model to be constructed;
and selecting factors from the factor tree according to the discriminant ability value as characteristic variables of the model.
An offence discriminating apparatus comprising:
the characteristic information acquisition module is used for acquiring characteristic information of the main body to be evaluated;
The scoring module is used for inputting the characteristic information into the offence risk scoring model to obtain offence risk scores;
the behavior record acquisition module is used for acquiring the behavior record of the main body to be evaluated in a set time window if the illegal risk score is larger than a set threshold value;
And the illegal behavior label output module is used for inputting the behavior record into a rule model to obtain an illegal behavior label.
In one embodiment, the apparatus further comprises:
and the risk pointing keyword mapping module is used for obtaining the risk pointing keywords corresponding to the evaluation main body according to the predefined association relationship between the illegal action tags and the risk pointing keywords.
In one embodiment, the apparatus further comprises:
The factor tree acquisition module is used for acquiring a factor tree corresponding to a main body to be evaluated, wherein nodes of the factor tree comprise factors describing characteristics of the main body to be evaluated;
The characteristic variable selection module is used for selecting factors from the factor tree as characteristic variables of the model according to target variables of the violation risk scoring model to be constructed;
And the violation risk score model construction module is used for constructing a violation risk score model according to the determined characteristic variable and the target variable.
A computer device comprising a memory storing a computer program and a processor implementing the steps of the method described above when the processor executes the computer program.
A computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the method described above.
According to the method, the device, the computer equipment and the storage medium for judging the illegal behaviors, the feature information of the main body to be evaluated is comprehensively analyzed through the illegal risk scoring model, and the illegal risk score is obtained; and screening a plurality of subjects to be evaluated according to the risk scores, and further evaluating the risk of the specific behavior record only for subjects with higher risk degree to obtain the illegal behavior label. The data screening is carried out by the rule model, the specific rule label output is carried out by the rule model, the calculation amount is reduced on the basis of guaranteeing the accuracy of the rule risk assessment, and the rule risk assessment efficiency is improved.
Drawings
FIG. 1 is an application scenario diagram of a method for offence determination in one embodiment;
FIG. 2 is a flow chart of a method for determining offensiveness according to an embodiment;
FIG. 3 is a schematic flow diagram of one embodiment involved in constructing a violation risk scoring model;
FIG. 4 is a schematic flow chart of another embodiment of a process involved in constructing a violation risk scoring model;
FIG. 5 is a block diagram of an apparatus for determining offensiveness in one embodiment;
FIG. 6 is a block diagram of the structures involved in constructing a violation risk scoring model in one embodiment;
fig. 7 is an internal structural diagram of a computer device in one embodiment.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
The method for judging the illegal behaviors can be applied to an application environment shown in figure 1. The application environment includes a terminal 102 and a server 104, wherein the terminal 102 and the server 104 may communicate over a network. The terminal 102 sends a risk assessment request to the server 104, the server 104 receives the risk assessment request, and extracts the identity of the subject to be assessed and the characteristic information of the subject to be assessed, which are carried in the request. And the server 104 calls a violation risk scoring model corresponding to the identity of the main body to be evaluated, and performs model analysis on the characteristic information of the main body to be evaluated through the violation risk scoring model to obtain a violation risk score. The server further screens out the subjects to be evaluated with the offence risk score larger than the set threshold value, only carries out scale model evaluation on the subjects to be evaluated which pass the screening, and outputs offence labels.
In one embodiment, as shown in fig. 2, there is provided a method for determining rule violations, which is illustrated by taking the server in fig. 1 as an example, and includes the following steps:
step 202: and acquiring the characteristic information of the main body to be evaluated.
The subject to be evaluated is a group with set attributes, such as staff, students, etc. The acquired characteristic information of the subject to be evaluated is the characteristic information of each individual to be evaluated in the acquisition group. Such as basic feature information, behavior feature information, capability feature information, emotional characteristic information, social characteristic information, and the like of the employee a.
Step 204: and inputting the characteristic information into the offence risk scoring model to obtain offence risk scores.
The offence risk scoring model is a supervised scoring model constructed according to a machine learning algorithm. Different offence risk scoring models correspond to different subjects. The scoring process in this embodiment is: and (3) calling a corresponding violation risk scoring model according to the main body identification to be evaluated in the step (202), acquiring an input characteristic variable of the called risk violation model, extracting characteristic information corresponding to the input characteristic variable from the acquired characteristic information of the main body to be evaluated, and inputting the extracted characteristic information into the violation risk scoring model to obtain the violation risk score of the main body to be evaluated.
Step 206: and if the illegal risk score is larger than the set threshold value, acquiring a behavior record of the main body to be evaluated in a set time window.
Judging whether the violation risk score output by the violation risk score model is larger than a set threshold, if yes, judging that the main body to be evaluated has high violation risk attribute. The server acquires the behavior record of the subject to be evaluated, which is judged to have the high risk attribute of the violation by the model, in the set time window. The time window of the acquired behavior record can be configured according to actual requirements.
If the subject to be evaluated is staff, the offence risk scoring model is an offence risk scoring model of staff, a path for acquiring data is preconfigured as a staff offence detail storage path, and the acquired behavior record is an offence behavior record of staff. For another example, if the subject to be evaluated is an employee, the offence risk scoring model is an attendance risk scoring model of the employee, the path for acquiring the data is preconfigured as an employee attendance storage path, and the acquired behavior records are the attendance records of the employee, the leave records and other behavior records related to attendance.
In one embodiment, an associated behavioral record data acquisition path is configured for each offence risk score model. When the behavior record of the main body to be evaluated is obtained, a data acquisition path associated with the illegal risk scoring model is obtained, and the behavior record of the main body to be evaluated is obtained according to the data acquisition path.
Step 208: and inputting the behavior record into the rule model to obtain the rule-breaking behavior label.
Inputting the obtained specific behavior record of the main body to be evaluated into a pre-constructed rule model, and carrying out rule violation judgment on the input behavior record by the rule model according to predefined rule violation judgment conditions, and outputting a plurality of rule violation labels. The method comprises the following steps: and predefining the rule violation judging conditions corresponding to each rule violation label, and outputting the corresponding rule violation label if the behavior record meets the rule violation judging conditions.
Taking staff illegal reimbursement as an example, if the illegal reimbursement behavior is that the single reimbursement amount is 5000000, the output illegal tags are as follows: the reimbursement amount is near (2%) the specified amount (500,000). If the violation is reimbursement invoice serial number, outputting the violation label as follows: and reimbursement of the invoice serial number for multiple times.
In this embodiment, a machine learning model, that is, a rule model is combined with a traditional rule model, and the machine learning model comprehensively analyzes multidimensional feature information of a subject to be evaluated to obtain a generalized rule risk evaluation result. And then acquiring the fine-grained risk embodiment data according to the generalized offence risk assessment result, and analyzing the risk embodiment data by the operation rule model to obtain a fine-grained offence risk assessment label. The former model is used for data screening, and the latter model is used for specific violation label output, so that the calculation amount is reduced on the basis of ensuring the accuracy of the violation risk assessment, and the violation risk assessment efficiency is improved.
In one embodiment, at step 208: after the behavior record is input into the rule model to obtain the rule tag, the method further comprises the following steps: and obtaining the risk pointing key corresponding to the evaluation subject according to the association relation between the predefined violation behavior label and the risk pointing key.
In terms of cost violation reimbursement, the risk-oriented keywords may be phrases with obvious offence risk orientations such as "reimbursement for a single reimbursement", "false invoice", "assault reimbursement for a reimbursement", and the like.
The rule-breaking label is essentially a standardized rule-breaking expression, and rule-breaking expressions in the professional field are generally rather unsmooth. In this embodiment, the association relationship between the rule violation label and the risk pointing keyword is predefined, and the rule violation label is mapped to the risk pointing keyword, so that the risk directivity of the rule violation evaluation result is more clear, and the output result is more visual and understandable.
For the rule-breaking risk scoring model in the above embodiment, as shown in fig. 3, a method for providing the rule-breaking risk scoring model specifically includes the following steps:
step 302: and obtaining a factor tree corresponding to the main body to be evaluated, wherein the nodes of the factor tree comprise factors describing the characteristics of the main body to be evaluated.
The factor tree is pre-constructed. In one embodiment, the factor tree is a tree structure formed by factors describing the characteristics of the subject to be evaluated, and the factor tree may be constructed by: and collecting factors for characterizing a plurality of dimension attribute of the main body, and marking the hierarchical labels of the collected factors. And connecting each factor according to the belonging level label to generate a tree structure diagram. The hierarchy labels to which factor a belongs are: the first branch prime factor, factor b, belongs to the hierarchy label: the first branch is from the factor, or from the sub-factor under the factor, and the sub-factor under the sub-factor, etc. In addition, the hierarchy attribute also contains other branches.
Further, each branch of the factor tree corresponds to an attribute of one dimension of the body. The branches of the subtrees and the nodes of each branch are increased and expanded without limitation. The subtrees should be as comprehensive as possible to reveal the factor composition of any dimension of the subject.
Step 304: and selecting factors from the factor tree as characteristic variables of the model according to target variables of the violation risk scoring model to be constructed.
And determining a target variable of the model to be constructed, and selecting factors from the factor tree according to the target variable to serve as characteristic variables of the model. And selecting a factor which can be the most different in the dimension of the target variable by the main body from the factor tree as a characteristic variable of the model.
For example, the model is constructed as a cost violation risk scoring model, the corresponding subject of the model is an "employee", and the determined target variable is the employee cost violation risk level. Correspondingly, the factor which can distinguish the high-risk illegal staff from the normal staff is selected from the factor tree to be used as the characteristic variable of the illegal risk scoring model.
In one embodiment, the discriminatory power value of each factor in the factor tree to the target variable of the violation risk scoring model to be constructed can be calculated through single factor analysis and correlation analysis; and selecting factors from the factor tree according to the discriminant ability value as characteristic variables of the model.
Step 306: and constructing a violation risk scoring model according to the determined characteristic variables and the target variables.
Sample data corresponding to the selected characteristic variables are collected, and a supervised learning mode is adopted to construct an offence risk scoring model.
In the embodiment, the model feature variables are selected from the factor tree containing all the features of the main body, so that the feature variables of the constructed model are more comprehensive, and the accuracy of model prediction is improved. And redundant model indexes with the same contribution are selected, namely, the selection of the model indexes is enabled to be in a vector based on a factor system.
In one embodiment, the factor tree is marked with data positioning information and data processing information of corresponding associated data for each factor. In this embodiment, as shown in fig. 4, the method for constructing the offence risk score model includes the following steps:
Step 402: and obtaining a factor tree corresponding to the main body to be evaluated, and selecting factors from the factor tree as characteristic variables of the model.
Step 404: data positioning information and data processing information associated with the feature variables of the model are obtained from the factor tree.
And labeling the data positioning information and the data processing information of the associated data corresponding to the factor under each factor node of the factor tree. The data location information may be a storage path for the associated data. The data processing information includes preprocessing algorithms, processing algorithms, etc. of the associated data.
In one embodiment, the corresponding degree of missing associated data may also be noted. When selecting the feature variable from the factor tree, a factor with a lower degree of deficiency may be preferentially selected as the feature variable of the model.
Step 406: and acquiring associated data corresponding to the characteristic variables of the model according to the data positioning information.
Step 408: and preprocessing the acquired associated data according to the data processing information to obtain a training sample.
Step 410: and inputting the training sample into a preselected model algorithm for supervised learning to obtain a violation risk scoring model.
And acquiring associated data corresponding to the feature variables according to the data positioning information, processing the acquired associated data according to the data processing information to obtain a training sample, and training the training sample to obtain the illegal risk scoring model.
It should be noted that, various models related to the corresponding subject may be constructed based on the factor tree, and are not limited to the above-described offence risk scoring model. Corresponding factors are selected from the factor tree to serve as characteristic variables according to the requirement of a specific construction model, and training data are obtained through data annotation in the factor tree, so that the required model can be constructed.
It should be understood that, although the steps in the flowcharts of fig. 2-4 are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in fig. 2-4 may include multiple sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, nor do the order in which the sub-steps or stages are performed necessarily occur sequentially, but may be performed alternately or alternately with at least a portion of the sub-steps or stages of other steps or steps.
In one embodiment, as shown in fig. 5, there is provided an offence determination apparatus, including:
the feature information obtaining module 502 is configured to obtain feature information of a subject to be evaluated.
And the scoring module 504 is used for inputting the characteristic information into the offence risk scoring model to obtain offence risk scores.
The behavior record obtaining module 506 is configured to obtain a behavior record of the subject to be evaluated in a set time window if the risk score of the violation is greater than a set threshold.
And the illegal action tag output module 508 is used for inputting the action record into the rule model to obtain the illegal action tag.
In one embodiment, the offence determination device further includes:
and the risk pointing keyword mapping module is used for obtaining the risk pointing keywords corresponding to the evaluation main body according to the association relation between the predefined violation behavior labels and the risk pointing keywords.
In one embodiment, as shown in fig. 6, the device for discriminating an offence further includes:
The factor tree obtaining module 602 is configured to obtain a factor tree corresponding to the subject to be evaluated, where a node of the factor tree includes a factor describing a feature of the subject to be evaluated.
The feature variable selection module 604 is configured to select a factor from the factor tree as a feature variable of the model according to a target variable of the offence risk score model to be constructed.
The offence risk score model construction module 606 is configured to construct an offence risk score model based on the determined feature variables and the target variables.
In one embodiment, each factor is marked with data positioning information and data processing information of corresponding associated data in the factor tree;
the offence risk score model construction module 606 is further configured to obtain data positioning information and data processing information associated with feature variables of the model from the factor tree; acquiring associated data corresponding to the characteristic variables of the model according to the data positioning information; preprocessing the acquired associated data according to the data processing information to obtain a training sample; and inputting the training sample into a preselected model algorithm for supervised learning to obtain a violation risk scoring model.
In one embodiment, the feature variable selection module 604 is further configured to calculate a discriminatory capability value of each factor in the factor tree for the target variable of the offence risk score model to be constructed; and selecting factors from the factor tree according to the discriminant ability value as characteristic variables of the model.
The specific limitation of the illegal activity discriminating device can be referred to the limitation of the illegal activity discriminating method hereinabove, and will not be described herein. The above-mentioned each module in the offence determination device may be implemented in whole or in part by software, hardware, and a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In one embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 7. The computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is for storing association data corresponding to factors in the factor tree. The network interface of the computer device is for communicating with the terminal through a network connection. The computer program, when executed by a processor, implements a method for determining offensiveness.
It will be appreciated by those skilled in the art that the structure shown in FIG. 7 is merely a block diagram of some of the structures associated with the present inventive arrangements and is not limiting of the computer device to which the present inventive arrangements may be applied, and that a particular computer device may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.
In one embodiment, a computer device is provided comprising a memory storing a computer program and a processor that when executing the computer program performs the steps of: acquiring characteristic information of a main body to be evaluated; inputting the characteristic information into a violation risk scoring model to obtain a violation risk score; if the illegal risk score is larger than the set threshold value, acquiring a behavior record of the main body to be evaluated in a set time window; and inputting the behavior record into the rule model to obtain the rule-breaking behavior label.
In one embodiment, the processor when executing the computer program further performs the steps of: and obtaining the risk pointing key corresponding to the evaluation subject according to the association relation between the predefined violation behavior label and the risk pointing key.
In one embodiment, the processor when executing the computer program further performs the steps of: acquiring a factor tree corresponding to a main body to be evaluated, wherein nodes of the factor tree comprise factors describing characteristics of the main body to be evaluated; selecting factors from the factor tree as characteristic variables of the model according to target variables of the violation risk scoring model to be constructed; and constructing a violation risk scoring model according to the determined characteristic variables and the target variables.
In one embodiment, each factor is marked with data positioning information and data processing information of corresponding associated data in the factor tree; the processor when executing the computer program also implements the steps of:
Constructing a violation risk scoring model according to the determined characteristic variables and the target variables, wherein the method comprises the following steps of: acquiring data positioning information and data processing information associated with characteristic variables of the model from the factor tree; acquiring associated data corresponding to the characteristic variables of the model according to the data positioning information; preprocessing the acquired associated data according to the data processing information to obtain a training sample; and inputting the training sample into a preselected model algorithm for supervised learning to obtain a violation risk scoring model.
In one embodiment, the processor when executing the computer program further performs the steps of: calculating the discrimination capability value of each factor in the factor tree for the target variable of the violation risk scoring model to be constructed; and selecting factors from the factor tree according to the discriminant ability value as characteristic variables of the model.
In one embodiment, a computer readable storage medium is provided having a computer program stored thereon, which when executed by a processor, performs the steps of: acquiring characteristic information of a main body to be evaluated; inputting the characteristic information into a violation risk scoring model to obtain a violation risk score; if the illegal risk score is larger than the set threshold, acquiring a behavior record of the main body to be evaluated in a set time window; and inputting the behavior records into the rule model to obtain one or more offence tags.
In one embodiment, the computer program when executed by the processor further performs the steps of: and obtaining the risk pointing key corresponding to the evaluation subject according to the association relation between the predefined violation behavior label and the risk pointing key.
In one embodiment, the computer program when executed by the processor further performs the steps of: acquiring a factor tree corresponding to a main body to be evaluated, wherein nodes of the factor tree comprise factors describing characteristics of the main body to be evaluated; selecting factors from the factor tree as characteristic variables of the model according to target variables of the violation risk scoring model to be constructed; and constructing a violation risk scoring model according to the determined characteristic variables and the target variables.
In one embodiment, each factor is marked with data positioning information and data processing information of corresponding associated data in the factor tree; the computer program when executed by the processor also performs the steps of:
Constructing a violation risk scoring model according to the determined characteristic variables and the target variables, wherein the method comprises the following steps of: acquiring data positioning information and data processing information associated with characteristic variables of the model from the factor tree; acquiring associated data corresponding to the characteristic variables of the model according to the data positioning information; preprocessing the acquired associated data according to the data processing information to obtain a training sample; and inputting the training sample into a preselected model algorithm for supervised learning to obtain a violation risk scoring model.
In one embodiment, the computer program when executed by the processor further performs the steps of: calculating the discrimination capability value of each factor in the factor tree for the target variable of the violation risk scoring model to be constructed; and selecting factors from the factor tree according to the discriminant ability value as characteristic variables of the model.
Those skilled in the art will appreciate that implementing all or part of the above-described methods may be accomplished by way of a computer program, which may be stored on a non-transitory computer readable storage medium and which, when executed, may comprise the steps of the above-described embodiments of the methods. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (SYNCHLINK) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
The technical features of the above-described embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above-described embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above examples illustrate only a few embodiments of the invention, which are described in detail and are not to be construed as limiting the scope of the invention. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the invention, which are all within the scope of the invention. Accordingly, the scope of protection of the present invention is to be determined by the appended claims.
Claims (10)
1. A method of offence discrimination, comprising:
Acquiring a factor tree corresponding to a main body to be evaluated, wherein nodes of the factor tree comprise factors describing characteristics of the main body to be evaluated;
Selecting factors from the factor tree as characteristic variables of the model according to target variables of the violation risk scoring model to be constructed;
Marking data positioning information and data processing information of corresponding associated data for each factor in the factor tree; acquiring the data positioning information and the data processing information corresponding to the characteristic variable from the factor tree; acquiring associated data corresponding to the characteristic variable according to the data positioning information; preprocessing the acquired associated data according to the data processing information to obtain a training sample; inputting the training sample into a preselected model algorithm for supervised learning to obtain a violation risk scoring model;
Acquiring characteristic information of a main body to be evaluated;
Invoking a violation risk scoring model corresponding to a to-be-evaluated subject identifier of the to-be-evaluated subject;
acquiring an input characteristic variable of the risk violation model; the input characteristic variables are factors selected from factor trees of the main body to be evaluated based on target variables of the risk violation model when the risk violation model is constructed, the factor trees are tree structure diagrams formed by factors describing the characteristics of the main body to be evaluated, and each branch in the factor trees respectively represents the attribute of one corresponding dimension of the main body to be evaluated;
Extracting feature information corresponding to the input feature variable from the feature information;
inputting the extracted characteristic information into the offence risk scoring model to obtain offence risk scores;
If the illegal risk score is larger than a set threshold value, acquiring a behavior record of the main body to be evaluated in a set time window;
And inputting the behavior record into a rule model, and carrying out rule violation judgment on the behavior record through rule violation judgment conditions corresponding to the rule model to obtain a rule violation label.
2. The method of claim 1, further comprising, after said entering the behavioral record into a rule model to obtain an offending behavior tag:
And obtaining the risk pointing key corresponding to the evaluation subject according to the association relation between the predefined illegal behavior label and the risk pointing key.
3. The method of claim 1, wherein the subject to be assessed is an employee, and the characteristic information includes basic characteristic information, behavioral characteristic information, capability characteristic information, emotional characteristic information, and social characteristic information.
4. The method of claim 1, wherein the data location information comprises a storage path of the associated data; the data processing information comprises a preprocessing algorithm and a processing algorithm of the associated data.
5. The method according to claim 3 or 4, wherein the selecting a factor from the factor tree as a model feature variable according to the target variable of the offence risk score model to be constructed comprises:
calculating the discrimination capability value of each factor in the factor tree for the target variable of the violation risk scoring model to be constructed;
and selecting factors from the factor tree according to the discriminant ability value as characteristic variables of the model.
6. An offence determination device, comprising:
The factor tree acquisition module is used for acquiring a factor tree corresponding to a main body to be evaluated, wherein nodes of the factor tree comprise factors describing characteristics of the main body to be evaluated;
The characteristic variable selection module is used for selecting factors from the factor tree as characteristic variables of the model according to target variables of the violation risk scoring model to be constructed;
The violation risk scoring model construction module is used for marking data positioning information and data processing information of corresponding associated data for each factor in the factor tree; acquiring the data positioning information and the data processing information corresponding to the characteristic variable from the factor tree; acquiring associated data corresponding to the characteristic variable according to the data positioning information; preprocessing the acquired associated data according to the data processing information to obtain a training sample; inputting the training sample into a preselected model algorithm for supervised learning to obtain a violation risk scoring model;
The characteristic information acquisition module is used for acquiring characteristic information of the main body to be evaluated; invoking a violation risk scoring model corresponding to a to-be-evaluated subject identifier of the to-be-evaluated subject; acquiring an input characteristic variable of the risk violation model; the input characteristic variables are factors selected from factor trees of the main body to be evaluated based on target variables of the risk violation model when the risk violation model is constructed, the factor trees are tree structure diagrams formed by factors describing the characteristics of the main body to be evaluated, and each branch in the factor trees respectively represents the attribute of one corresponding dimension of the main body to be evaluated; extracting feature information corresponding to the input feature variable from the feature information;
The scoring module is used for inputting the extracted characteristic information into the offence risk scoring model to obtain offence risk scores;
the behavior record acquisition module is used for acquiring the behavior record of the main body to be evaluated in a set time window if the illegal risk score is larger than a set threshold value;
and the rule-breaking behavior label output module is used for inputting the behavior record into a rule model, and performing rule-breaking judgment on the behavior record through rule-breaking judgment conditions corresponding to the rule model to obtain a rule-breaking behavior label.
7. The apparatus of claim 6, wherein the apparatus further comprises:
and the risk pointing keyword mapping module is used for obtaining the risk pointing keywords corresponding to the evaluation main body according to the predefined association relationship between the illegal action tags and the risk pointing keywords.
8. The apparatus of claim 6, wherein the subject to be assessed is an employee, and the characteristic information includes basic characteristic information, behavioral characteristic information, capability characteristic information, emotional characteristic information, and social characteristic information.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any one of claims 1 to 5 when the computer program is executed.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910068622.0A CN109886554B (en) | 2019-01-24 | 2019-01-24 | Illegal behavior discrimination method, device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910068622.0A CN109886554B (en) | 2019-01-24 | 2019-01-24 | Illegal behavior discrimination method, device, computer equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109886554A CN109886554A (en) | 2019-06-14 |
| CN109886554B true CN109886554B (en) | 2024-05-03 |
Family
ID=66926782
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910068622.0A Active CN109886554B (en) | 2019-01-24 | 2019-01-24 | Illegal behavior discrimination method, device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109886554B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111324370B (en) * | 2020-02-19 | 2022-10-18 | 支付宝(杭州)信息技术有限公司 | Method and device for carrying out risk processing on to-be-on-line small program |
| CN111369369A (en) * | 2020-03-16 | 2020-07-03 | 中科天玑数据科技股份有限公司 | System and method for monitoring microdisk transaction platform |
| CN111754044B (en) * | 2020-06-30 | 2024-04-30 | 深圳前海微众银行股份有限公司 | Employee behavior auditing method, device, equipment and readable storage medium |
| CN112417459B (en) * | 2020-11-19 | 2022-10-28 | 上海浦东发展银行股份有限公司 | Large-scale terminal equipment safety assessment method and system and computer equipment |
| CN113407727B (en) * | 2021-03-22 | 2023-01-13 | 天津汇智星源信息技术有限公司 | Qualitative measure and era recommendation method based on legal knowledge graph and related equipment |
| CN113327162B (en) * | 2021-06-29 | 2022-03-08 | 深圳市魔数智擎人工智能有限公司 | Credit wind control rule making system |
| CN115543391B (en) * | 2022-11-29 | 2023-04-04 | 深圳市明源云科技有限公司 | Application platform upgrading method and device, electronic equipment and readable storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106446270A (en) * | 2016-10-19 | 2017-02-22 | 广东欧珀移动通信有限公司 | Classifying method and device |
| CN107492033A (en) * | 2017-08-30 | 2017-12-19 | 广东信基蜂巢科技有限责任公司 | A kind of credit estimation method and device based on air control model |
| CN108132981A (en) * | 2017-12-13 | 2018-06-08 | 国家电网公司 | Take the electric power facility site selecting method of Rainstorm Flood venture influence into account |
| CN108427669A (en) * | 2018-02-27 | 2018-08-21 | 华青融天（北京）技术股份有限公司 | Abnormal behaviour monitoring method and system |
| CN108427708A (en) * | 2018-01-25 | 2018-08-21 | 腾讯科技（深圳）有限公司 | Data processing method, device, storage medium and electronic device |
| CN108876600A (en) * | 2018-08-20 | 2018-11-23 | 平安科技（深圳）有限公司 | Warning information method for pushing, device, computer equipment and medium |
| CN109165849A (en) * | 2018-08-27 | 2019-01-08 | 众安信息技术服务有限公司 | Methods of risk assessment and device |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090248560A1 (en) * | 2008-03-28 | 2009-10-01 | Fortent Americas Inc. | Assessment of risk associated with doing business with a party |
| US8548934B2 (en) * | 2009-11-17 | 2013-10-01 | Infozen, Inc. | System and method for assessing risk |
-
2019
- 2019-01-24 CN CN201910068622.0A patent/CN109886554B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106446270A (en) * | 2016-10-19 | 2017-02-22 | 广东欧珀移动通信有限公司 | Classifying method and device |
| CN107492033A (en) * | 2017-08-30 | 2017-12-19 | 广东信基蜂巢科技有限责任公司 | A kind of credit estimation method and device based on air control model |
| CN108132981A (en) * | 2017-12-13 | 2018-06-08 | 国家电网公司 | Take the electric power facility site selecting method of Rainstorm Flood venture influence into account |
| CN108427708A (en) * | 2018-01-25 | 2018-08-21 | 腾讯科技（深圳）有限公司 | Data processing method, device, storage medium and electronic device |
| CN108427669A (en) * | 2018-02-27 | 2018-08-21 | 华青融天（北京）技术股份有限公司 | Abnormal behaviour monitoring method and system |
| CN108876600A (en) * | 2018-08-20 | 2018-11-23 | 平安科技（深圳）有限公司 | Warning information method for pushing, device, computer equipment and medium |
| CN109165849A (en) * | 2018-08-27 | 2019-01-08 | 众安信息技术服务有限公司 | Methods of risk assessment and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109886554A (en) | 2019-06-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109886554B (en) | Illegal behavior discrimination method, device, computer equipment and storage medium | |
| WO2020253358A1 (en) | Service data risk control analysis processing method, apparatus and computer device | |
| CN108876600B (en) | Early warning information pushing method, device, computer equipment and medium | |
| CN109165840B (en) | Risk prediction processing method, risk prediction processing device, computer equipment and medium | |
| CN109272396B (en) | Customer risk early warning method, device, computer equipment and medium | |
| TWI723528B (en) | Computer-executed event risk assessment method and device, computer-readable storage medium and computing equipment | |
| CN111767707B (en) | Method, device, equipment and storage medium for detecting Leideogue cases | |
| CN109829629B (en) | Risk analysis report generation method, apparatus, computer device and storage medium | |
| CN109767322A (en) | Suspicious transaction analysis method, apparatus and computer equipment based on big data | |
| CN109543925B (en) | Risk prediction method and device based on machine learning, computer equipment and storage medium | |
| CN110490720A (en) | Financial data analysis and early warning method, apparatus, computer equipment and storage medium | |
| CN106776884B (en) | Terrorism prediction method based on multi-label combination and multi-classifier | |
| CN110458324B (en) | Method and device for calculating risk probability and computer equipment | |
| CN109801151B (en) | Financial falsification risk monitoring method, device, computer equipment and storage medium | |
| CN109767326A (en) | Suspicious transaction reporting generation method, device, computer equipment and storage medium | |
| CN112580329B (en) | Text noise data identification method, device, computer equipment and storage medium | |
| CN112990989B (en) | Value prediction model input data generation method, device, equipment and medium | |
| CN112464660B (en) | Text classification model construction method and text data processing method | |
| CN114328942A (en) | Relationship extraction method, apparatus, device, storage medium and computer program product | |
| CN116991455A (en) | API asset identification method and device | |
| CN114692785B (en) | Behavior classification method, device, equipment and storage medium | |
| CN110909975A (en) | Scientific research platform benefit evaluation method and device | |
| CN115310606A (en) | Deep learning model depolarization method and device based on data set sensitive attribute reconstruction | |
| CN111401055B (en) | Method and apparatus for extracting context information from financial information | |
| CN113591480A (en) | Named entity identification method and device for power metering and computer equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |