CN115081922A - New risk active identification method and device based on map library - Google Patents
New risk active identification method and device based on map library Download PDFInfo
- Publication number
- CN115081922A CN115081922A CN202210802345.3A CN202210802345A CN115081922A CN 115081922 A CN115081922 A CN 115081922A CN 202210802345 A CN202210802345 A CN 202210802345A CN 115081922 A CN115081922 A CN 115081922A
- Authority
- CN
- China
- Prior art keywords
- risk
- detection model
- candidate
- determining
- target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 57
- 238000001514 detection method Methods 0.000 claims abstract description 138
- 238000001228 spectrum Methods 0.000 claims abstract description 17
- 238000004590 computer program Methods 0.000 claims description 16
- 238000003860 storage Methods 0.000 claims description 14
- 238000012549 training Methods 0.000 claims description 11
- 238000000605 extraction Methods 0.000 claims description 10
- 230000000007 visual effect Effects 0.000 claims description 10
- 230000004044 response Effects 0.000 claims description 3
- 230000009286 beneficial effect Effects 0.000 abstract description 3
- 238000004891 communication Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 230000006399 behavior Effects 0.000 description 4
- 230000007547 defect Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000012800 visualization Methods 0.000 description 2
- 206010000117 Abnormal behaviour Diseases 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000002860 competitive effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 210000003462 vein Anatomy 0.000 description 1
Images
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
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Development Economics (AREA)
- Quality & Reliability (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- General Business, Economics & Management (AREA)
- Educational Administration (AREA)
- Operations Research (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a new risk active identification method and device based on a map library. The method comprises the following steps: establishing a detection model according to a pre-established knowledge map library and a modeling platform; wherein the detection model comprises at least one of a rule detection model, an auto-supervised learning detection model and an unsupervised learning detection model; determining candidate risk objects through the detection model and the knowledge spectrum library; and judging whether a target risk object and/or a target risk group exist or not according to the object information of the candidate risk object, and determining a risk identification result according to a judgment result. By operating the technical scheme provided by the embodiment of the invention, the problem that the existing risks and the risks which are possibly generated cannot be effectively identified can be solved, and the beneficial effects of improving the accuracy and efficiency of risk identification are achieved.
Description
Technical Field
The invention relates to a computer technology, in particular to a new risk active identification method and device based on a map library.
Background
At present, thousands of small and medium-sized enterprises exist in China, and how to provide effective financial services for the small and medium-sized enterprises is an important subject facing each large financial institution. In the service process, the financial institution faces the difficulties of how to identify potential good customers and how to screen risk customers.
Therefore, a risk identification method is needed to help financial institutions better evaluate the liability of a client and reduce the potential financial risk, so as to help financial institutions solve various problems in the development process of the general finance.
Disclosure of Invention
The invention provides a new risk active identification method and device based on a map library, and aims to improve the accuracy and efficiency of risk identification.
According to one aspect of the invention, a new risk active identification method based on a map library is provided, which is characterized by comprising the following steps:
establishing a detection model according to a pre-established knowledge map library and a modeling platform; wherein the detection model comprises at least one of a rule detection model, an auto-supervised learning detection model and an unsupervised learning detection model;
determining candidate risk objects through the detection model and the knowledge spectrum library;
and judging whether a target risk object and/or a target risk group exist or not according to the object information of the candidate risk object, and determining a risk identification result according to a judgment result.
According to another aspect of the present invention, there is provided a new risk active identification device based on a map library, comprising:
the detection model establishing module is used for establishing a detection model according to a pre-established knowledge map library and a modeling platform; wherein the detection model comprises at least one of a rule detection model, an auto-supervised learning detection model and an unsupervised learning detection model;
a candidate risk object determination module for determining candidate risk objects through the detection model and the knowledge spectrum library;
and the risk identification result determining module is used for judging whether a target risk object and/or a target risk group exist according to the object information of the candidate risk object and determining a risk identification result according to a judgment result.
According to another aspect of the present invention, there is provided an electronic apparatus including:
at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the method for active recognition of new risks based on a atlas database according to any of the embodiments of the present invention.
According to another aspect of the present invention, there is provided a computer-readable storage medium storing computer instructions for causing a processor to implement the method for active recognition of new risk based on a atlas database according to any embodiment of the present invention when the computer instructions are executed.
According to the technical scheme of the embodiment of the invention, a detection model is established according to a pre-established knowledge map library and a modeling platform; wherein the detection model comprises at least one of a rule detection model, an auto-supervised learning detection model and an unsupervised learning detection model; determining candidate risk objects through the detection model and the knowledge spectrum library; and judging whether a target risk object and/or a target risk group exist or not according to the object information of the candidate risk object, and determining a risk identification result according to a judgment result. The problem that existing risks and risks which are possibly generated cannot be effectively identified is solved, and the beneficial effects of improving accuracy and efficiency of risk identification are achieved.
It should be understood that the statements in this section are not intended to identify key or critical features of the embodiments of the present invention, nor are they intended to limit the scope of the invention. Other features of the present invention will become apparent from the following description.
Drawings
Fig. 1 is a flowchart of a new risk active identification method based on a spectrum library according to an embodiment of the present invention;
fig. 2 is a flowchart of a new risk active identification method based on a atlas database according to a second embodiment of the present invention;
fig. 3 is a schematic structural diagram of a new risk active identification system based on a spectrum library according to a second embodiment of the present invention;
fig. 4 is a schematic structural diagram of a new risk active identification apparatus based on a chart library according to a third embodiment of the present invention;
fig. 5 is a schematic structural diagram of an electronic device for implementing an embodiment of the invention.
Detailed Description
In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, shall fall within the protection scope of the present invention.
It should be noted that the terms "first," "second," "target," and the like in the description and claims of the present invention and in the above-described drawings are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
Example one
Fig. 1 is a flowchart of a new risk active identification method based on a spectrogram library according to an embodiment of the present invention, which is applicable to a situation where risk objects are identified based on a knowledge graph, and the method can be executed by a new risk active identification device based on a spectrogram library according to an embodiment of the present invention, and the device can be implemented in a software and/or hardware manner. Referring to fig. 1, the new risk active identification method based on a atlas database provided in this embodiment includes:
s110, establishing a detection model according to a pre-constructed knowledge map library and a modeling platform; wherein the detection model comprises at least one of a regular detection model, an auto-supervised learning detection model, and an unsupervised learning detection model.
The pre-constructed knowledge map library can integrate, store and extract known basic information and behavior characteristic data of the object to be detected in real time, such as archive information, behavior activity data, digital fingerprints, inter-account relation, threshold optimization and the like. And deducing fine-grained and rich information, thereby improving the subsequent detection effect of the detection model.
The modeling platform can be a visualization platform and is used for providing one-stop combined modeling service for sample import, data matching, feature processing, model training, model evaluation and other operations. Therefore, the real-time performance, effectiveness and rapidness of the detection model establishment are ensured.
The method comprises the following steps of establishing a rule detection model through real-time updated detection rules provided by a modeling platform and high-dimensional features of data in a database extracted from a knowledge spectrum database; establishing a self-supervision learning detection model through model construction rules of a machine learning model provided by a modeling platform and extracting high-dimensional characteristics of data in a database from a knowledge spectrum library; and establishing an unsupervised learning detection model through model construction rules of the machine learning model provided by the modeling platform and extracting high-dimensional characteristics of data in the database from the knowledge spectrum library.
The unsupervised learning detection model does not need labels and training data, can prevent in advance and flexibly cope with continuously evolving risk modes, and can detect in advance, for example, when a risk object is a risk account, the risk account can be detected when the risk account is applied or registered.
The unsupervised learning detection model can input and output information from the knowledge spectrum library, and summarize and calculate the information of a plurality of digital fingerprints. And new and unknown potential risk objects or high-quality objects are effectively screened out, so that great competitive advantage is brought to financial institutions.
The rule detection model belongs to the application of expert experience based on a rule strategy, the self-supervision learning detection model has higher marking requirement on a data sample, and the two methods are effective in detecting known risk types but have defects in detecting unknown risks. The unsupervised learning detection model can make up the defects of the rule detection model and the self-supervised learning detection model.
In this embodiment, optionally, the establishing a detection model according to a pre-constructed knowledge graph library and a modeling platform includes:
determining a feature extraction mode according to data information of data in the knowledge map library;
and extracting the data characteristics of the data in the knowledge map library according to the characteristic extraction mode, and establishing the detection model according to the data characteristics.
The data information may include the data type and data field length of the data. Different data characteristics can be extracted according to different data types, for example, when the data is transaction amount, the data characteristics of a digital class are extracted; and when the data is the user label, extracting the data characteristics of the label class, thereby extracting the applicable characteristics as much as possible in real time. When the data field of the data changes or increases, the number of the extracted feature dimensions can be automatically adjusted correspondingly.
And dynamically adjusting corresponding feature extraction modes according to different data to improve the effectiveness of feature extraction, thereby extracting the data features of the data in the corresponding knowledge map library, establishing a detection model according to the data features and improving the accuracy of model establishment.
And S120, determining candidate risk objects through the detection model and the knowledge spectrum library.
And determining candidate risk objects corresponding to the detection models through at least one detection model and the knowledge spectrum library. For example, a first set of candidate risk objects is determined by a rule detection model, a second set of candidate risk objects is determined by an unsupervised learning detection model, and a third set of candidate risk objects is determined by an unsupervised learning detection model. The type of the candidate risk object may be predetermined, for example, a financial account, and the like, which is not limited in this embodiment.
Data which are not subjected to feature extraction in the knowledge map library provide data support for operation of the detection model. For example, if a first candidate risk object is detected by the detection model, more comprehensive related information of the first candidate risk object or information of other objects having an association relationship with the first candidate risk object is obtained from the knowledge map library to find other candidate risk objects.
And detecting the total data or the questionable data through a rule detection model, and determining the objects meeting the risk object rules in the total data or the questionable data provided by the knowledge map library as candidate risk objects.
The self-supervision learning detection model takes the behavior mode of the detected risk object as model training data, and determines the similar detected risk object in the total data or the suspicious data provided by the knowledge map library as the candidate risk object.
The rule detection model and the self-supervision learning detection model need to be detected by depending on certain standards, and can start from 6 dimensions such as identity information, time, space, quantity, association, labels and the like, and start from 72 veins such as account opening information, label information, transaction regions, transaction time, transaction frequency, transaction association, transaction purposes and the like, 360-degree scanning such as fund source going direction, access characteristics, supervision avoidance, intermittent transaction and the like is performed, so that omnibearing stereoscopic perspective of a detected object is realized, a candidate risk object determination result is given, and the accuracy of candidate risk object determination is improved.
By "unsupervised" is meant that the abnormal behavior pattern is not known before, nor are those that are valid features known. Because there is no presetting, the features can be viewed in a high-dimensional space and extracted comprehensively. Candidate risk objects are determined from the unlabeled data using clustering/graph structure analysis techniques that focus on studying the relationships and connections between data provided by the knowledge-graph library. Illustratively, new suspected activities and risk behaviors can be automatically discovered by analyzing the distances and connections between data points in a knowledge graph library and analyzing a graph structure formed by the data points, wherein the data points represent accounts and the activities of the accounts within a certain time.
S130, judging whether a target risk object and/or a target risk group exist according to the object information of the candidate risk object, and determining a risk identification result according to a judgment result.
For example, when the candidate risk object is a financial account, the object information may be transaction frequency of the financial account, and the like, which is not limited in this embodiment.
When a single candidate risk object is detected in different detection models, information about the candidate risk object can be fused to finally obtain a unique candidate risk object, so that the accuracy of determining the candidate risk object and the comprehensiveness of the obtained object information are improved.
Whether the target risk object exists is judged according to the object information of the candidate risk object, which can be used for judging the degree that the object information meets the requirement of the target risk object, for example, when various object information exists, the number that the object information meets the requirement of the target risk object is judged, and if the number is larger than the required threshold value, the candidate risk object is determined to be the target risk object.
Judging whether target risk groups exist according to the object information of the candidate risk objects, namely judging the relevance between different candidate risk objects according to the object information of the candidate risk objects, if a candidate risk object set with higher relevance exists, judging the degree that the set information of the candidate risk object set meets the requirements of the target risk groups, for example, judging the number of set information meeting the requirements of the target risk groups when various set information exists, and if the number is greater than a required threshold value, determining that the candidate risk object set is the target risk groups.
And determining a risk identification result according to the judgment result, and summarizing the judgment result to obtain a risk identification result. For example, the risk identification result is the existing target risk object and/or target risk group, and the risk degree of the target risk object and/or target risk group, etc., which is not limited in this embodiment.
Optionally, the judgment result and/or the risk identification result is added to the knowledge graph library and/or the modeling platform, so that the knowledge graph library and/or the modeling platform are continuously updated in an iterative manner, and during each iteration, the selected features, weights and distance functions are automatically adjusted until the selected features, weights and distance functions meet the relevant standards, so that the reliability of data in the knowledge graph library and/or the modeling platform is improved.
According to the technical scheme provided by the embodiment, a detection model is established according to a pre-constructed knowledge map library and a modeling platform; wherein the detection model comprises at least one of a rule detection model, an automatic supervision learning detection model and an unsupervised learning detection model; determining candidate risk objects through the detection model and the knowledge spectrum library; and judging whether a target risk object and/or a target risk group exist according to the object information of the candidate risk object, and determining a risk identification result according to a judgment result. The method has the advantages that comprehensive, three-dimensional and dynamic evaluation is effectively carried out on the detected objects, so that the method can be used for obtaining the risk degree of the detected objects and screening high-quality objects from the detected objects; for example, the method can effectively identify black production information hidden in the network and can also identify potential good customers in the network. The mechanism is assisted to rapidly improve the product wind control and operation capacity in the aspects of client recommendation, group fraud, black intermediary identification, complex wind control rule realization and the like. The problem that existing risks and risks which are possibly generated cannot be effectively identified is solved, and the effects of improving accuracy and efficiency of risk identification are achieved.
Example two
Fig. 2 is a flowchart of a new risk active identification method based on a atlas database according to a second embodiment of the present invention, and this technical solution is described in an additional way for a process of determining whether a target risk object and/or a target risk group exists according to the object information of the candidate risk object. Compared with the scheme, the scheme is specifically optimized to judge whether the target risk object and/or the target risk group exist or not according to the object information of the candidate risk object, and comprises the following steps:
determining a credibility score of the candidate risk object according to the object information of the candidate risk object;
sorting the candidate risk objects according to the credibility scores, and determining whether the target risk objects exist according to the sorting result;
and determining whether target risk groups exist or not according to the object information of the target risk object. Specifically, a flow chart of the new risk active identification method based on the atlas database is shown in fig. 2:
s210, establishing a detection model according to a pre-constructed knowledge map library and a modeling platform; wherein the detection model comprises at least one of a regular detection model, an auto-supervised learning detection model, and an unsupervised learning detection model.
And S220, determining candidate risk objects through the detection model and the knowledge spectrum library.
And S230, determining the credibility score of the candidate risk object according to the object information of the candidate risk object.
The credibility score can be carried out based on the condition that the object information of the candidate risk object meets the preset expert rule. And setting different weights and scores according to different rules, summarizing the satisfied conditions, and obtaining the total credibility score of each candidate risk object.
S240, sorting the candidate risk objects according to the credibility scores, and determining whether the target risk objects exist according to the sorting result.
The candidate risk objects are ranked according to the confidence score, and whether a target risk object exists is determined according to the ranking result, which may be that the candidate risk object in a preset ranking range is determined as the target risk object, or that the candidate risk object with the confidence score greater than or equal to the preset confidence threshold is determined as the target risk object according to a preset confidence threshold, which is not limited in this embodiment.
And S250, determining whether a target risk group exists according to the object information of the target risk object.
Determining whether target risk groups exist according to the object information of the target risk objects, judging the relevance between different target risk objects according to the object information of the target risk objects, if a target risk object set with higher relevance exists, judging the degree that the set information of the target risk object set meets the requirements of the target risk groups, for example, judging the number of the set information meeting the requirements of the target risk groups when multiple sets of information exist, and if the number is greater than a required threshold value, determining that the target risk object set is the target risk groups.
In this embodiment, optionally, determining whether a target risk group exists according to the object information of the target risk object includes:
determining candidate risk groups according to the object information of the target risk object;
and determining whether the target risk group exists according to the group information of the candidate risk group.
And analyzing the object information of the target risk object through a visual graph, and clustering and associating the similar or strongly associated target risk objects to obtain candidate risk groups. The group information of the candidate risk groups can comprise attack properties of the groups, group scale and the like, and the group candidate risk groups can be classified according to the attack properties of the groups, so that each candidate risk group classification can be scored by using a preset function based on the related group scale and the corresponding clustering distance according to the classification. The smaller the natural clustering distance, the larger the ganged scale and the higher the score, whether the target risk ganged exists is determined according to the scoring result, and the accuracy of determining the target risk ganged is improved.
And S260, determining a risk identification result according to the judgment result.
In this embodiment, optionally, the method further includes:
determining sub-risk objects of the candidate risk objects according to the unsupervised learning detection model and/or the rule detection model;
and training the self-supervision learning detection model according to the sub-risk object.
And the sub-risk objects are candidate risk objects respectively obtained by an unsupervised learning detection model or a rule detection model. And inputting the sub-risk object into the self-supervision learning detection model to serve as a training sample of the self-supervision learning detection model, and expanding the training sample, so that the accuracy of establishing the self-supervision learning detection model is improved.
In this embodiment, optionally, the method further includes:
and updating the rule detection model according to the operation result of the self-supervision learning detection model.
And automatically adjusting part of parameters in the detection model according to the operation result of the self-supervision learning detection model, for example, whether the operation result meets the preset requirement or not, so that the operation result of the self-supervision learning detection model obtained by training the sub-risk object determined by the rule detection model meets the preset requirement. The rule detection model automatically changes rules on the basis of keeping the transparency of the rule system, and the effectiveness of the rule detection model is improved.
In this embodiment, optionally, the method further includes:
and displaying and/or changing the risk identification result in response to an operation request of a user through a visual interface.
Based on the visual analysis technology, the incidence relation, the scoring result and the like of the target risk object are displayed in a visual interface in a visual mode, so that the examiner can conveniently and visually and quickly process the complex result, the user experience is improved, the processing result can be updated to the knowledge map library and the modeling platform, and the real-time performance of data in the knowledge map library and the modeling platform is improved.
Fig. 3 is a schematic structural diagram of a new risk active recognition system based on a spectrogram library according to a second embodiment of the present invention, as shown in fig. 3, a detection model is established according to a modeling platform and data features extracted from data in a knowledge spectrogram library; wherein the detection model comprises at least one of a rule detection model, an automatic supervision learning detection model and an unsupervised learning detection model; determining respective candidate risk objects through the detection model and the knowledge graph library; determining the credibility score of the candidate risk object according to the object information of the candidate risk object; the candidate risk objects are sorted according to the credibility scores, and whether the target risk objects exist is determined according to the sorting result; and determining whether target risk groups exist according to the target information of the target risk object, and determining a risk identification result according to a judgment result. And displaying and/or changing the risk identification result in response to the operation request of the user through the visual interface.
The candidate risk object ranking results and the visualization analysis results may be fed back to the modeling platform and from the knowledge graph repository to update the modeling platform and from the knowledge graph repository. Determining sub-risk objects in the candidate risk objects according to the unsupervised learning detection model and/or the rule detection model; and training a self-supervision learning detection model according to the sub-risk objects. And updating the rule detection model according to the operation result of the self-supervision learning detection model. By utilizing the technologies of knowledge maps, artificial intelligence, visual analysis and the like, the comprehensive scanning recognition of rule discrimination, supervised learning, unsupervised learning and the like is realized through core functions of self-researched rule detection models, unsupervised learning detection, unsupervised learning, modeling platforms and the like, and the risk of customers can be effectively evaluated. Not only can the known customer model be identified, but also the new unknown customer risk can be automatically learned. The system provides a series of capabilities such as identity recognition, anti-fraud, credit assessment, risk monitoring and the like for institutions such as banks and internet finance, and helps financial institutions to solve various problems in the development process of the conventional finance.
According to the embodiment of the invention, the credibility score of the candidate risk object is determined according to the object information of the candidate risk object; the candidate risk objects are sorted according to the credibility scores, and whether the target risk objects exist is determined according to the sorting result; and determining whether the target risk group exists or not according to the target information of the target risk object, and improving the accuracy of determining the target risk object and/or the target risk group.
EXAMPLE III
Fig. 4 is a schematic structural diagram of a new risk active identification device based on a atlas database according to a third embodiment of the present invention. The device can be realized in a hardware and/or software mode, can execute the new risk active identification method based on the map library provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method. As shown in fig. 4, the apparatus includes:
the detection model establishing module 410 is used for establishing a detection model according to a pre-established knowledge map library and a modeling platform; wherein the detection model comprises at least one of a rule detection model, an auto-supervised learning detection model and an unsupervised learning detection model;
a candidate risk object determination module 420 for determining candidate risk objects from the detection model and the knowledge spectrum library;
and a risk identification result determining module 430, configured to determine whether a target risk object and/or a target risk group exists according to the object information of the candidate risk object, and determine a risk identification result according to a determination result.
On the basis of the above technical solutions, optionally, the detection model establishing module includes:
the characteristic extraction mode determining unit is used for determining a characteristic extraction mode according to data information of data in the knowledge map library;
and the detection model establishing unit is used for extracting the data characteristics of the data in the knowledge graph library according to the characteristic extraction mode and establishing the detection model according to the data characteristics.
On the basis of the above technical solutions, optionally, the risk identification result determining module includes:
a credibility score determining unit, configured to determine a credibility score of the candidate risk object according to object information of the candidate risk object;
an object existence determining unit, configured to rank the candidate risk objects according to the credibility score, and determine whether the target risk object exists according to the ranking result;
and the group existence determining unit is used for determining whether the target risk group exists according to the object information of the target risk object.
On the basis of the above technical solutions, optionally, the group presence determining unit includes:
a group determining subunit, configured to determine candidate risk groups according to the object information of the target risk object;
and the group existence determining subunit is used for determining whether the target risk group exists according to the group information of the candidate risk group.
On the basis of the above technical solutions, optionally, the apparatus further includes:
a sub-risk object determination module for determining sub-risk objects of the candidate risk objects according to the unsupervised learning detection model and/or the rule detection model;
and the detection model training module is used for training the self-supervision learning detection model according to the sub-risk objects.
On the basis of the above technical solutions, optionally, the apparatus further includes:
and the rule detection model updating module is used for updating the rule detection model according to the operation result of the self-supervision learning detection model.
On the basis of the above technical solutions, optionally, the apparatus further includes:
and the result display and change module is used for responding to an operation request of a user through a visual interface and displaying and/or changing the risk identification result.
Example four
FIG. 5 illustrates a schematic diagram of an electronic device 10 that may be used to implement an embodiment of the invention. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the inventions described and/or claimed herein.
As shown in fig. 5, the electronic device 10 includes at least one processor 11, and a memory communicatively connected to the at least one processor 11, such as a Read Only Memory (ROM)12, a Random Access Memory (RAM)13, and the like, wherein the memory stores a computer program executable by the at least one processor, and the processor 11 can perform various suitable actions and processes according to the computer program stored in the Read Only Memory (ROM)12 or the computer program loaded from a storage unit 18 into the Random Access Memory (RAM) 13. In the RAM 13, various programs and data necessary for the operation of the electronic apparatus 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other via a bus 14. An input/output (I/O) interface 15 is also connected to bus 14.
A number of components in the electronic device 10 are connected to the I/O interface 15, including: an input unit 16 such as a keyboard, a mouse, or the like; an output unit 17 such as various types of displays, speakers, and the like; a storage unit 18 such as a magnetic disk, an optical disk, or the like; and a communication unit 19 such as a network card, modem, wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information/data with other devices via a computer network such as the internet and/or various telecommunication networks.
The processor 11 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various processors running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, or the like. The processor 11 performs the various methods and processes described above, such as a new risk active identification method based on a gallery.
In some embodiments, the atlas-based active identification of new risks method may be implemented as a computer program tangibly embodied in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and/or installed onto the electronic device 10 via the ROM 12 and/or the communication unit 19. When loaded into RAM 13 and executed by processor 11, the computer program may perform one or more steps of the atlas library-based active identification of new risks method described above. Alternatively, in other embodiments, the processor 11 may be configured to perform a atlas database-based new risk active identification method by any other suitable means (e.g., by means of firmware).
Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuitry, Field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), system on a chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be special or general purpose, receiving data and instructions from, and transmitting data and instructions to, a storage system, at least one input device, and at least one output device.
A computer program for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions/acts specified in the flowchart and/or block diagram block or blocks to be performed. A computer program can execute entirely on a machine, partly on a machine, as a stand-alone software package partly on a machine and partly on a remote machine or entirely on a remote machine or server.
In the context of the present invention, a computer-readable storage medium may be a tangible medium that can contain, or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. A computer readable storage medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, the computer readable storage medium may be a machine readable signal medium. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
To provide for interaction with a user, the systems and techniques described here can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which a user can provide input to the electronic device. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic, speech, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a back-end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), Wide Area Networks (WANs), blockchain networks, and the internet.
The computing system may include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so that the defects of high management difficulty and weak service expansibility in the traditional physical host and VPS service are overcome.
It should be understood that various forms of the flows shown above may be used, with steps reordered, added, or deleted. For example, the steps described in the present invention may be executed in parallel, sequentially, or in different orders, and are not limited herein as long as the desired results of the technical solution of the present invention can be achieved.
The above-described embodiments should not be construed as limiting the scope of the invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may be made in accordance with design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A new risk active identification method based on a map library is characterized by comprising the following steps:
establishing a detection model according to a pre-established knowledge map library and a modeling platform; wherein the detection model comprises at least one of a rule detection model, an auto-supervised learning detection model and an unsupervised learning detection model;
determining candidate risk objects through the detection model and the knowledge spectrum library;
and judging whether a target risk object and/or a target risk group exist or not according to the object information of the candidate risk object, and determining a risk identification result according to a judgment result.
2. The method of claim 1, wherein building the detection model from a pre-built knowledge map library and a modeling platform comprises:
determining a feature extraction mode according to data information of data in the knowledge map library;
and extracting the data characteristics of the data in the knowledge map library according to the characteristic extraction mode, and establishing the detection model according to the data characteristics.
3. The method of claim 1, wherein determining whether a target risk object and/or a target risk group exists based on the object information of the candidate risk objects comprises:
determining a credibility score of the candidate risk object according to the object information of the candidate risk object;
sorting the candidate risk objects according to the credibility scores, and determining whether the target risk objects exist according to the sorting result;
and determining whether target risk groups exist or not according to the object information of the target risk object.
4. The method of claim 3, wherein determining whether a target risk group exists based on the subject information of the target risk subject comprises:
determining candidate risk groups according to the object information of the target risk object;
and determining whether the target risk group exists according to the group information of the candidate risk group.
5. The method of claim 1, further comprising:
determining sub-risk objects of the candidate risk objects according to the unsupervised learning detection model and/or the rule detection model;
and training the self-supervision learning detection model according to the sub-risk object.
6. The method of claim 5, further comprising:
and updating the rule detection model according to the operation result of the self-supervision learning detection model.
7. The method of claim 1, further comprising:
and displaying and/or changing the risk identification result in response to an operation request of a user through a visual interface.
8. A new risk active recognition device based on a map library is characterized by comprising:
the detection model establishing module is used for establishing a detection model according to a pre-established knowledge map library and a modeling platform; wherein the detection model comprises at least one of a rule detection model, an auto-supervised learning detection model and an unsupervised learning detection model;
a candidate risk object determination module for determining candidate risk objects through the detection model and the knowledge spectrum library;
and the risk identification result determining module is used for judging whether a target risk object and/or a target risk group exist according to the object information of the candidate risk object and determining a risk identification result according to a judgment result.
9. An electronic device, characterized in that the electronic device comprises:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein,
the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the atlas database-based new risk active identification method of any of claims 1-7.
10. A computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions for causing a processor to implement the atlas-based new risk active identification method of any of claims 1-7 when executed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210802345.3A CN115081922A (en) | 2022-07-07 | 2022-07-07 | New risk active identification method and device based on map library |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210802345.3A CN115081922A (en) | 2022-07-07 | 2022-07-07 | New risk active identification method and device based on map library |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115081922A true CN115081922A (en) | 2022-09-20 |
Family
ID=83258146
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210802345.3A Pending CN115081922A (en) | 2022-07-07 | 2022-07-07 | New risk active identification method and device based on map library |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115081922A (en) |
-
2022
- 2022-07-07 CN CN202210802345.3A patent/CN115081922A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110287316A (en) | A kind of Alarm Classification method, apparatus, electronic equipment and storage medium | |
| KR102113347B1 (en) | Method, apparatus and computer program for classifying cryptocurrency accounts using artificial intelligence | |
| CN112950359B (en) | User identification method and device | |
| CN115619245A (en) | Portrait construction and classification method and system based on data dimension reduction method | |
| CN114186626A (en) | Abnormity detection method and device, electronic equipment and computer readable medium | |
| WO2024098699A1 (en) | Entity object thread detection method and apparatus, device, and storage medium | |
| CN114692778A (en) | Multi-modal sample set generation method, training method and device for intelligent inspection | |
| CN117235608B (en) | Risk detection method, risk detection device, electronic equipment and storage medium | |
| CN110930242A (en) | Credibility prediction method, device, equipment and storage medium | |
| CN115965464A (en) | Empty shell enterprise identification method and device, storage medium and electronic device | |
| CN115630708A (en) | Model updating method and device, electronic equipment, storage medium and product | |
| CN115081922A (en) | New risk active identification method and device based on map library | |
| CN114818892A (en) | Credit grade determining method, device, equipment and storage medium | |
| CN114943608A (en) | Fraud risk assessment method, device, equipment and storage medium | |
| CN114444514A (en) | Semantic matching model training method, semantic matching method and related device | |
| CN113052512A (en) | Risk prediction method and device and electronic equipment | |
| CN114511022A (en) | Feature screening, behavior recognition model training and abnormal behavior recognition method and device | |
| CN114155589A (en) | Image processing method, device, equipment and storage medium | |
| CN113807391A (en) | Task model training method and device, electronic equipment and storage medium | |
| CN111429257A (en) | Transaction monitoring method and device | |
| CN116244740B (en) | Log desensitization method and device, electronic equipment and storage medium | |
| CN112949752B (en) | Training method and device of business prediction system | |
| CN117609723A (en) | Object identification method and device, electronic equipment and storage medium | |
| CN115392399A (en) | Method, device, equipment and medium for training and using process timeout prediction model | |
| CN116467198A (en) | Method, device, electronic equipment and storage medium for determining performance actual measurement necessity |
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 |