CN111311136A - Wind control decision method, computer equipment and storage medium - Google Patents
Wind control decision method, computer equipment and storage medium Download PDFInfo
- Publication number
- CN111311136A CN111311136A CN202010405458.0A CN202010405458A CN111311136A CN 111311136 A CN111311136 A CN 111311136A CN 202010405458 A CN202010405458 A CN 202010405458A CN 111311136 A CN111311136 A CN 111311136A
- Authority
- CN
- China
- Prior art keywords
- data
- wind control
- user
- portrait
- decision
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- 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
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
-
- 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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Finance (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- General Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Game Theory and Decision Science (AREA)
- Databases & Information Systems (AREA)
- Marketing (AREA)
- Educational Administration (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- General Engineering & Computer Science (AREA)
- Information Transfer Between Computers (AREA)
Abstract
The application relates to the technical field of internet information, and discloses a wind control decision method, computer equipment and a storage medium, wherein the method comprises the following steps: receiving active data and structured data from a client, wherein the active data comprises data submitted by a user, the structured data comprises data obtained by structuring interactive data and/or passive data by the client, the interactive data comprises user operation actions, and the passive data comprises client information; verifying the active data, analyzing the structured data, and storing the verified active data and the analyzed data obtained by analysis in a database; generating user characteristic data according to the verified active data and the analyzed data based on the portrait model in the portrait model library so as to establish a user portrait of the current user; generating a wind control result according to the user portrait based on a wind control decision engine; and responding the service request of the client according to the wind control result. The wind control decision can be preferably realized based on big data, machine learning and the like.
Description
Technical Field
The present application relates to the technical field of internet information technologies, and in particular, to a wind control decision method, a computer device, and a storage medium.
Background
With the rise and development of the internet, various services based on the internet come out endlessly. Due to the unstable credibility of the internet information, for most internet services, the core competitiveness of the internet is influenced by the internet wind control.
The internet has been born so far, and people have thought that information from the internet is far less reliable than information from people directly interacting with the middle ears and seeing and hitting. Therefore, most internet practices, which involve the wind control link, are still human-oriented. Taking the current internet finance as an example, the mainstream of the wind control includes real-time investigation, third-party deposit of funds, examination of the flow of funds, and payment of risk preparation funds. These four general approaches may exist completely excluding internet technology.
Future internet services, including internet finance, are credit-based services. Such credit, based on a trusted record, builds on widespread citation, has a very low transaction barrier and a very high production efficiency, and is the mainstream in the future.
With the development of internet technology and the emergence of advanced technologies such as cloud computing, big data, artificial intelligence and the like, information with high-quality trusted components in internet information can be extracted through comprehensive processing with certain calculation power, and the information with the credibility far exceeding that of the traditional so-called reality can be extracted.
However, the wind control system in the internet industry is still at a lower level at present, the manual execution is more, the automation level is low, and the wind control effect is poor. How to solve the problems becomes a key for improving the overall social benefit of the internet business.
Disclosure of Invention
The embodiment of the application provides a wind control decision method, computer equipment and a storage medium, which can better realize the wind control decision based on big data, machine learning and the like.
In a first aspect, the present application provides a method for wind control decision, where the method includes:
receiving active data and structured data from a client, wherein the active data comprises data submitted by a user, the structured data comprises data obtained by the client performing structured processing on interactive data and/or passive data, the interactive data comprises user operation actions, and the passive data comprises client information;
verifying the active data, analyzing the structured data, and storing the verified active data and the analyzed data obtained by analysis in a database;
generating user characteristic data according to the verified active data and the analyzed data based on the portrait model in the portrait model library so as to establish a user portrait of the current user;
generating a wind control result according to the user portrait based on a wind control decision engine;
responding the service request of the client according to the wind control result;
wherein the wind control decision engine comprises at least one decision tree comprising at least one node;
the wind control decision engine is used for generating a wind control result according to the user portrait, and comprises the following steps:
generating a corresponding wind control variable according to corresponding user characteristic data based on a decision node of the decision tree, and adjusting a wind control coefficient of the decision tree according to the wind control variable;
and generating a wind control result according to the wind control coefficient of at least one decision tree.
In a second aspect, the present application provides a method for wind control decision, including:
determining whether the current user is a bad machine user, and rejecting the bad machine user;
the method comprises the steps of obtaining original data of a current user, wherein the original data comprise at least one of active data, interactive data and passive data, the active data comprise data submitted by the user, the interactive data comprise user operation actions, and the passive data comprise client information;
verifying the active data and sending the verified active data to a server;
carrying out structuring processing on the interactive data and/or the passive data to obtain structured data;
the method comprises the steps of sending structured data to a server based on a preset uploading rule, verifying the active data and analyzing the structured data by the server, generating user feature data according to the verified active data and the analyzed data to obtain a user portrait of a current user based on a portrait model in a portrait model library, generating a wind control result according to the user portrait by the server based on a wind control decision engine, and responding to a service request of a client according to the wind control result.
In a third aspect, the present application provides a computer device comprising a memory and a processor; the memory is used for storing a computer program; the processor is used for executing the computer program and realizing the wind control decision method when the computer program is executed.
In a fourth aspect, the present application provides a computer-readable storage medium, where a computer program is stored, and if the computer program is executed by a processor, the method for wind control decision-making is implemented.
The application discloses a wind control decision method, equipment and a storage medium, wherein active data and structured data are received from a client, the active data comprise data submitted by a user, the structured data comprise data obtained by the client performing structured processing on interactive data and/or passive data, the interactive data comprise user operation actions, and the passive data comprise client information; verifying the active data, analyzing the structured data, and storing the verified active data and the analyzed data obtained by analysis in a database; generating user characteristic data according to the verified active data and the analyzed data based on the portrait model in the portrait model library so as to establish a user portrait of the current user; generating a wind control result according to the user portrait based on a wind control decision engine; and responding the service request of the client according to the wind control result. The method comprises the steps of collecting and organizing original data locally by utilizing the limited available computing power of a client, packaging the data into structured data, uploading the data collected for multiple times to a server, rapidly spreading the data on the server, obtaining a user image through an image data processing algorithm, outputting results of fraud detection and risk assessment through a wind control decision engine, and feeding back the results to a business logic response client. By using the method and the system, a client model and an algorithm can be prevented from being probed, an output result is prevented from being falsified to influence a wind control decision result of a server, the client can be prevented from purposefully modifying original data to cause pollution of a server database and disordered model training, on the other hand, network bandwidth, system calculation and storage pressure can be greatly reduced through pre-clearing of bad bugs, and user experience of real-time application is improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a schematic flow chart of a wind control decision method according to an embodiment of the present application;
FIG. 2 is a schematic view of an application scenario of a wind control decision method;
FIG. 3 is a schematic diagram of a decision node;
FIG. 4 is a schematic diagram of a logical structure of a wind control decision engine;
FIG. 5 is a schematic diagram of a logical structure for generating new rule nodes;
FIG. 6 is a schematic diagram of a logical structure for generating new scoring nodes;
fig. 7 is a schematic flow chart of a wind control decision method according to another embodiment of the present application;
fig. 8 is a schematic structural diagram of a computer device according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some, but not all, embodiments of the present application. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The flow diagrams depicted in the figures are merely illustrative and do not necessarily include all of the elements and operations/steps, nor do they necessarily have to be performed in the order depicted. For example, some operations/steps may be decomposed, combined or partially combined, so that the actual execution sequence may be changed according to the actual situation.
Fig. 1 is a schematic flow chart of a wind control decision method according to an embodiment of the present application. The wind control decision method can be applied to a terminal or a server to realize fraud detection and/or risk control on a service request of a client.
For example, the wind control decision method is used for a server, and of course, may be used for a terminal. Wherein, the terminal can be electronic equipment such as a computer; the servers may be independent servers or server clusters. However, for the sake of understanding, the following embodiments will be described in detail with reference to a wind control decision method applied to a server.
As shown in fig. 2, the server and client are communicatively coupled. The client is configured with a corresponding browser and/or APP, the browser and/or APP running can send a service request to the server, and the server process can respond to the service request.
Illustratively, the business request includes at least one of internet financial services such as a network credit application, a payment application, a financing application, and a financial insurance purchase. For example, the user's application of Html5 at the client triggers the client to send a service request to the server.
The H5 application can support the access of the foot bugs (bots), can fully exert the propagation characteristics of the Internet and has strong information popularization efficiency. As a good wind control method and application, supporting H5 will become an important market demand. Native APP can be avoided because by native program parcel, form the data isolated island, lack the agreeable nature with the internet, the not enough of visibility is low.
For example, the wind control decision method of the present embodiment is mainly applied to a scenario where the client runs the H5 application. The client is not required to have high computational power and storage support, so that the application in a browser H5 environment can be supported.
Some embodiments of the present application will be described in detail below with reference to the accompanying drawings. The embodiments described below and the features of the embodiments can be combined with each other without conflict.
Referring to fig. 1, fig. 1 is a schematic flow chart of a wind control decision method according to an embodiment of the present application.
As shown in fig. 1, the wind control decision method includes the following steps S110 to S150.
Step S110, active data and structured data are received from the client.
The active data comprises data submitted by a user, the structured data comprises data obtained by a client performing structured processing on interactive data and/or passive data, the interactive data comprises user operation actions, and the passive data comprises client information.
In some embodiments, the client determines whether the current user is a bad machine user, and rejects the bad machine user if the current user is determined to be a bad machine user.
Over 90% of H5 application accesses on the internet are accesses from worms, such as web sniffers, web analyzers, content scrapers, brute force intrusion tools, search engine crawlers, etc., which are more frequent and faster than normal human accesses, not only presenting a security risk, but also occupying a lot of network bandwidth, computing and storage resources. Therefore, the rejection of bad machine users, such as dangerous and useless bugs, is an important safeguard to improve system safety and operating efficiency. On the other hand, beneficial insects can be released, so that the popularization and the propagation of the power-assisted information can be facilitated.
For example, partial calculation and storage of the client may be utilized to determine whether the current user is a bad machine user according to a preset machine identification logic, for example, the client determines whether the current user is a bad machine user according to a stored black list and a white list. By not allowing a bad machine user to submit requests and data to the server through the client, system safety can be improved, and resource waste such as bandwidth, storage and computing power can be reduced.
Illustratively, a bug (bot) refers to a non-human visitor, a malicious bug refers to a bug visitor with a system or business purpose that is compromised, such as a violence breaker, a leak sniffer, a malicious water-pouring bug and the like, and a useless bug refers to a bug that does not have a system or business purpose that is compromised but causes occupation of system resources, such as a content scraper, an automatic query tool and the like.
In some embodiments, the client obtains raw data of a current user that is not determined to be a bad machine user, the raw data including at least one of active data, interactive data, passive data.
For example, active data is data actively submitted by a user through a form, and is generally related to business, such as personal data submitted by the user, uploaded pictures, sent information, and the like; the interactive data is data generated by the operation and interaction between the user and the application system, such as a webpage browsed by the user, a viewed picture, a clicked button, a pull-down scroll bar and the like; the passive data is static information such as the environment in which the user is located, for example, the IP address of the user, the operating system used, the size of the screen, the model of the device, and the like.
Illustratively, the client verifies the active data and sends the verified active data to the server. And the client carries out structuring processing on the interactive data and/or the passive data to obtain structured data, and the client sends the structured data to the server based on a preset uploading rule.
Specifically, active data collected by the client is verified by the client, if the active data does not pass the verification, an error is returned to a user interface of the client to prompt an error, and if the active data passes the verification, the active data is directly uploaded to the server.
Illustratively, the processing the interaction data and/or the passive data in a structured manner includes: and at least one of classification, packaging, compression, temporary storage and combination is carried out on the interactive data and/or the passive data.
Illustratively, the client processes the collected interactive data and passive data, classifies them into categories, and organizes them into an ordered structured data system. In the structured data system, the structure, the form, the position and the like of the data are all provided with information, so that the information content (negative entropy value) of the data with the same length is improved, and the storage, the transmission, the extraction, the analysis and the management are facilitated. For example, user positioning data is organized into timestamps, IP, and GPS locations. Wherein, the GPS position comprises longitude, latitude, altitude, speed, direction and accuracy; the IP comprises an IP address and a geographic position resolved by the IP address, wherein the geographic position comprises a country, a city, a network provider, an IP longitude and an IP latitude; the time stamps are used to time-order the data of the different time samples.
For example, in a structure, when the geographical location of the post-analysis to the GPS and the geographical location of the IP resolution are very different, it may be determined that the user may use the VPN network or the proxy server.
Illustratively, the client may serialize the structured data for storage in the client database. When new homogeneous data is acquired, the client can extract a related data structure body from the client database, attach the newly acquired data to the client database, and store the newly acquired data back to the client database for storage after the newly acquired data is re-serialized.
Specifically, the client uploads the structured data in the client database to the server under the condition specified by the client code rule. It is also possible to leave an upload record locally and then clean up locally already uploaded structured data to free up local storage space.
Illustratively, the preset upload rule includes: and sending the structured data to the server according to the data volume of the structured data, the last time of sending the structured data and/or whether a key event is detected.
For example, when the structured data recorded by the client reaches a certain number, such as 1000 pieces, the structured data is sent to the server; or the structured data recorded by the client reaches a certain capacity, such as 1M byte, and the structured data is sent to the server; or the time from last uploading is longer than a certain time, such as 1 month, and the structured data recorded in the time period is sent to the server; or the client side sends the structured data to the server when detecting the occurrence of a key event, such as successful registration.
The client-side is used for not influencing the partial computing power and storage of experience, the access of bad insects is eliminated, the user data which do not violate the privacy right is collected, the active data is uploaded to the server in real time, and the interactive and passive data are classified, packaged, cached, accumulated and compressed locally and uploaded to the server according to the uploading rule.
And step S120, verifying the active data, analyzing the structured data, and storing the verified active data and the analyzed data obtained by analysis in a database.
And the server receives the active data uploaded by the client, verifies the validity of the data, and stores the verified active data into the database. After the active data uploaded to the server by the client reaches the server, the active data is verified by the server, service processing can be performed according to the verified active data, and the active data is stored in the database. The reason why the verification is carried out again is that the verification at the client is non-final, and because the client program is held in the hands of the client, technically, a user can tamper with the verification or skip the verification step by a very means, so that the aim of submitting illegal data is fulfilled. The verification of the client side is used for off-line verification, network requirements are reduced, verification speed is improved, user operation is guided in real time, and user experience is improved. The verification of the server belongs to final verification, and the safety of the data before entering the database is ensured.
And the server receives the structured data uploaded by the client and expands the structured data into a wind control data object. By analyzing the structured data, the obtained analyzed data is more suitable for database access and query. Illustratively, the parsing the structured data includes: extracting, sorting, compressing, counting, format converting, marking and the like to the structured data, and storing the processed analytic data into a database.
By way of example, storing validated active data in a NoSQL database can provide efficient short-term fast storage.
And S130, generating user characteristic data according to the verified active data and the analyzed data based on the portrait model in the portrait model library so as to establish the user portrait of the current user.
Illustratively, a representation model is used to build a user representation, which is a combination of data relevant to some aspect of the user, and its derived quantifiable or classifiable high-level features. User characteristic data, such as derivative features of the user representation, may be determined based on the representation model, for example. In particular, derived features of the representation can be obtained through feature definition, statistical data analysis (sometimes in conjunction with external data), and big data machine learning optimization.
Specifically, user characteristic data can be generated according to active data, interactive data and passive data of a user, and a user portrait can be determined according to various user characteristic data. For example, the server calculates user characteristic data of the user, such as a portrait derived characteristic value, from the real-time data of the user according to the portrait model in the portrait model library, namely, a portrait characteristic model algorithm, establishes a real-time portrait for the user, and periodically stores the real-time portrait in the database.
In some implementations, these data can be utilized to build a user representation of several dimensions.
Illustratively, the portrait model in the portrait model library may include at least one of a device portrait model, a character portrait model, a behavior portrait model, a business portrait model, a social portrait model; at least one of a user's equipment portrait, a character portrait, a behavior portrait, a business portrait and a social portrait can be correspondingly established.
Illustratively, a device representation of the current user is determined from the passive data based on a device representation model in a representation model library, the device representation including at least one of a device profile, a size-crowd level, a super-guest level, a usage circle category. For example, a user device portrait is created according to information of a device used by the user, including a device type, a device model, an operating system type, an operating system version, a network card number, a screen size, whether root, installed software and the like, and derived features such as a device grade, a size and mass level, a super class, a use circle class and the like are extracted through a device portrait model algorithm. Such information and features may provide rules and criteria for vector windmilling.
Illustratively, a portrait of the current user is determined from active data of the current user based on a character portrait model in a portrait model library, the portrait including at least one of traits, preferences, personality, value orientation, and goals. For example, a person portrait of the user is established according to the information of the user, including name, age, sex, language, country, race, height, weight and the like, and derived features such as traits, favorites, characters, value orientation, targets and the like are extracted through a person portrait model algorithm. Such information and features may provide rules and criteria for vector windmilling.
Illustratively, a behavior representation of the current user is determined according to the interaction data based on behavior representation models in a representation model library, wherein the behavior representation comprises at least one of office family or family, network dependency, taste and interest, and action style. For example, a behavior portrait of the user is established according to the behavior information of the user using the application system, including a daily use time period, frequently accessed pages, frequently viewed pictures and videos, a mouse click and typing ratio, a mouse click speed, a typing speed, a browsing speed and the like, and derived features such as office family or family, network dependency, taste and interest, a behavior style and the like are extracted through a behavior portrait model algorithm. Such information and features may provide rules and criteria for vector windmilling.
Illustratively, based on the business portrait model in the portrait model library, the business portrait of the current user is determined according to the business information of the current user, wherein the business portrait comprises at least one of customer loyalty, customer viscosity, customer impression and customer interaction liveness. For example, a business figure of the user is established according to the information related to the business provided by the user using the application system, such as the monthly expenditure of the user in the e-commerce system, the monthly times of purchase of the user, the monthly goods types of the user, whether the user frequently feeds back the evaluation and the like, and the derived characteristics of the customer loyalty, the customer viscosity, the customer impression, the customer interaction activity and the like are extracted through a business figure model algorithm. Such information and features may be applied to business design and management.
Illustratively, based on a social representation model in a representation model library, a social representation of the current user is determined according to the social relationship information of the current user, wherein the social representation comprises at least one of user social strength, user social breadth, user social depth and user social status. For example, a social image of the user, such as the number of contacts, the number of telephone calls and short messages per day, which apps are installed, and the like, is established according to the social relationship information of the user, and derived features such as the social strength of the user, the social breadth of the user, the social depth of the user, the social status of the user and the like are extracted through a social image model algorithm. Such information and features may be applied to business design and marketing.
It should be understood that the above description of the portrait type defined by the portrait model, the contained portrait type, and the related raw data and the extractable user feature data are only provided as some examples, and are not intended to limit the embodiments of the present disclosure.
In some embodiments, the method further comprises: and acquiring third-party data and/or historical data of the current user.
Illustratively, the third-party data includes data obtained from credit card platforms, e-commerce platforms, social network platforms, operator platforms, social security service platforms, accumulation fund service platforms, banks, and the like.
Illustratively, the historical data of the current user includes, for example, historical transaction data of the current user, and the like.
Illustratively, the user characteristic data is generated based on the portrait model in the portrait model library according to the analytic data and the third party data and/or historical data to establish the user portrait of the current user.
In some embodiments, the method further comprises: and deleting, updating or adding the portrait model in the portrait model library.
Illustratively, the deleting, updating or adding the portrait model in the portrait model library includes: training a portrait model according to portrait labels of a plurality of users and active data and/or the analytic data based on a machine learning algorithm; evaluating the effect of the portrait model, and updating or adding the portrait model which is evaluated to the portrait model library. The portrait label may be manually marked, for example.
Specifically, a better portrait model, namely a portrait feature algorithm, can be obtained through centralized integration and statistics of related original data information and combination of third-party data through machine learning and training. For example, for obtaining the extreme characteristics in the device image, the corresponding image model can be established by learning which device information is related to the extreme, such as whether root is available, technical software installed, and the like, and the degree of the correlation, parameterizing the related device information, and giving weight. The image model, namely the image characteristic algorithm is stored in the image model base after the effect meets the requirement. So as to generate user feature data based on the portrait model in the portrait model library according to the validated active data and the parsed data to create the user portrait of the current user in step S130.
Illustratively, as a preferred embodiment of the machine learning model training and evolution for the portrait model of the present invention, the training process of the portrait model machine learning model includes the following steps S301 to S305.
Step S301, the server acquires the original data of the client user collected and uploaded by each client.
Step S302, the server analyzes the original data and stores the original data into a database associated with the user according to the image types.
Step S303, the server uses massive user portrait data (or third party data can also be used) in the database and corresponding business result data to train the machine learning model in an off-line manner.
And step S304, the server predicts a new image instance by using the model, compares the service result after a period of time, evaluates the model effect and improves the model effect.
In step S305, the server adds the validated machine learning model to the model library, or updates the outdated model, and may delete the validated model from the image model library.
And step S140, generating a wind control result according to the user portrait based on a wind control decision engine.
In particular, the wind control decision engine may be used for fraud detection and risk assessment.
In some embodiments, the wind control decision engine comprises at least one decision tree comprising at least one node.
Illustratively, the wind control decision engine can synthesize data of all aspects, make decision rules, apply different decision trees, make comprehensive decisions, and output results of fraud detection and risk assessment.
Illustratively, the generating a wind control result from the user profile based on a wind control decision engine includes: generating a corresponding wind control variable according to corresponding user characteristic data based on a decision node of the decision tree, and adjusting a wind control coefficient of the decision tree according to the wind control variable; and generating a wind control result according to the wind control coefficient of at least one decision tree.
Illustratively, each decision tree at least comprises one wind control rule, each wind control rule is a decision node of the decision tree, and each decision node outputs at least one wind control variable by combining with user characteristic data generated by at least one portrait model. The wind control variables comprise positive wind control variables and negative wind control variables. Before the wind control result is obtained, each decision tree at least comprises one wind control coefficient, and the wind control variable output by each decision node is added to the wind control coefficient to change the value of the wind control variable. And when the wind control coefficient is smaller than the preset threshold value, the decision flows to a positive result, and the wind control decision engine integrates the wind control coefficients output by the decision trees and outputs the final results of fraud detection and risk assessment.
Illustratively, the wind control decision engine, which generates a wind control result according to the user portrait, further includes: and jumping to the next decision node according to the corresponding user characteristic data based on the decision node of the decision tree.
Illustratively, the wind control decision engine, which generates a wind control result according to the user portrait, further includes: and generating a wind control result according to the corresponding user characteristic data based on the decision nodes of the decision tree.
In some embodiments, the decision nodes comprise rule nodes and/or score nodes. The rule node comprises a plurality of wind control rules and combinational logic of the plurality of wind control rules, and the scoring node comprises a plurality of scoring factors and respective weights of the scoring factors and the combinational logic of the scoring factors.
Illustratively, the generating the corresponding wind control variable according to the corresponding user characteristic data based on the decision node of the decision tree includes: and generating corresponding wind control variables according to corresponding user characteristic data based on the rule nodes and/or the scoring nodes.
Illustratively, as shown in fig. 3, the decision nodes include a hard rule node 1, a hard rule node 2, a soft rule node 3, a score node 4, and a risk factor node 5.
The hard rule node 1 obtains two results according to the corresponding user characteristic data, directly refuses the results if the results are judged to be yes, terminates the audit, and continues the next audit if the results are judged to be no; the hard rule node 2 obtains two results according to the corresponding user characteristic data, if the results are judged to be yes, the results pass directly, the audit is terminated, and if the results are judged to be no, the next audit is continued; the soft rule node 3 obtains two results according to the corresponding user characteristic data, if the result is positive, the risk coefficient of the decision tree is increased or decreased according to the additional risk variable value defined by the rule, the next examination is continued, and if the result is negative, the next examination is directly continued; the scoring node 4 comprises a group of scoring factors with weights, the risk variable output is obtained through an internal algorithm, the risk variable is output after the scoring node 4 is applied to increase and decrease the risk coefficient of the decision tree, and the next step of auditing is continued; the risk coefficient node 5 may obtain a final decision result of the decision tree by performing value determination on the risk coefficient at the decision tree terminal. For example, a decision to reject is output when the risk factor exceeds 80, a decision to pass is output when the risk factor is less than 20, and a decision to submit a manual review is output when the risk factor is between 20 and 80.
And S150, responding to the service request of the client according to the wind control result.
Illustratively, the server responds to the service request of the client to realize fraud detection and/or risk control on the service request of the client.
In some embodiments, as shown in fig. 4, a schematic logical structure diagram of the wind control decision engine is shown, which illustrates execution logic and related components of the wind control decision engine.
Illustratively, the parameterized logic manager 430 selects a required rule node from the rule group library 440, and the number of the selected rule nodes may be multiple; the parameterization logic manager 430 selects nodes to be scored from the scoring card library 450, and the number of the selected scoring nodes can be multiple; and logically combining the selected rule nodes and the score nodes to establish a complete wind control decision engine 470, and storing the complete wind control decision engine 470 in a wind control decision engine model library 460. When the server responds to the service request of the client, the server selects one wind control decision engine 470 from the wind control decision engine model library 460 as the current wind control decision engine 470, and the current wind control decision engine 470 generates a wind control result, i.e., a decision 490, according to the service request 480.
In some embodiments, the method further comprises: and determining a plurality of wind control rules and the combinational logic of the plurality of wind control rules from a rule base based on a machine learning algorithm and/or the management operation of wind control personnel to generate a new rule node.
Illustratively, fig. 5 is a schematic diagram of a logic structure for generating a new rule node, which illustrates relevant components of the rule node.
Illustratively, the construction of new rule nodes may be done by the machine learning 510 or the parameterized logic manager 580, as follows: select one or more desired rules from rules repository 520 to join current rule set 530, see selected rule 1, selected rule 2,. the selected rule n; logic combination is carried out on the selected rules by using logic elements in the combinational logic 540, and a rule node is established; the created rule node is saved to the rule set library 550.
Specifically, the machine learning 510 uses a machine learning method to select the required rules and combinational logic; the parameterized logic manager 580 sets these rules and logic using a mechanism of manual intervention 590.
Illustratively, the rules in rules repository 520 may be derived from user information or derived features in the user representation, but may also be derived from other parameters.
The combinational logic 540 includes combinational logic elements of: AND, OR, NOT, XOR, GROUP. The GROUP is a combination element, which can combine two or more rules and then perform logic operation with other rules or rule combinations.
Manual intervention 590 is an intervention from an external human. For example, professional wind control personnel manually select rules and combinational logic through a parameterized logic manager through a manual intervention interface, set algorithm logic and store the algorithm logic in a rule set library. The manual intervention can establish a brand-new rule node and can also be modified into a new version on the basis of the rule node of machine learning.
In some embodiments, the method further comprises: and determining a plurality of scoring factors, the weights of the scoring factors and the combination logic of the scoring factors from a scoring factor library based on the management operation of a machine learning algorithm and/or a wind control person to generate a new scoring node.
Illustratively, fig. 6 is a schematic diagram of a logic structure for generating a new scoring node, which illustrates relevant components of the scoring node.
Illustratively, the construction of new scoring nodes may be done by machine learning 610 or parameterized logic manager 680 as follows: selecting one or more required scoring factors from the scoring factor library 620 to add to the current scoring card 630, see selected factor 1, selected factor 2,. the selected factor n; using the logic elements in the combination logic 640 to logically combine the selected scoring factors to establish a scoring node; the created scoring nodes are saved to the scoring card library 650.
Specifically, the machine learning 610 uses a machine learning method to select a desired scoring factor and a combination logic; the parameterization logic manager 680 uses the mechanism of manual intervention 690 to set these scoring factors and logic.
The scoring factors in the scoring factor library 620 may be derived from user information or derived features in the user representation, for example, but may be derived from other parameters.
The combinational logic 640 includes combinational logic elements having: AND, OR, NOT, XOR, GROUP. The GROUP is a combination element, which can combine two or more scoring factors and then perform logic operation with other scoring factors or the combination of the scoring factors.
Manual intervention 690 is an intervention from an external human. For example, a professional wind control person manually selects the scoring factors and the combination logic through a parameterized logic manager through a manual intervention interface, sets the algorithm logic, and stores the algorithm logic in the scoring card library 650. The manual intervention can establish a brand-new scoring node and can also be modified into a new version on the basis of the machine learning scoring node.
The machine learning has the advantages that the model is optimized comprehensively, meticulously and reliably and has strong improvement capability by utilizing big data and the big computing power of a computer; the disadvantages are that sufficient and comprehensive data support is needed, otherwise the model is seriously failed, and the learning speed is slow. The advantage of manual intervention is that a feasible model can be quickly obtained under the condition of lacking data by virtue of human knowledge, experience and intuition; the disadvantages are incomplete, not delicate, easily influenced by personal characteristics, etc. Therefore, the accuracy and the reliability of the wind control decision engine can be improved by integrating the advantages of the wind control decision engine and the wind control decision engine.
In the wind control decision method provided by the above embodiment, active data and structured data are received from a client, where the active data includes data submitted by a user, the structured data includes data obtained by performing a structured processing on interactive data and/or passive data by the client, the interactive data includes a user operation action, and the passive data includes client information; verifying the active data, analyzing the structured data, and storing the verified active data and the analyzed data obtained by analysis in a database; generating user characteristic data according to the verified active data and the analyzed data based on the portrait model in the portrait model library so as to establish a user portrait of the current user; generating a wind control result according to the user portrait based on a wind control decision engine; and responding the service request of the client according to the wind control result. The method comprises the steps of utilizing limited available computing power of a client and storing locally acquired and organized original data, packaging the data into structured data, uploading the data acquired for multiple times to a server, rapidly spreading the data on the server, obtaining user images through an image data processing algorithm, outputting results of fraud detection and risk assessment through a wind control decision engine, and feeding back the results to a business logic response client. By using the method and the system, a client model and an algorithm can be prevented from being probed, an output result is prevented from being falsified to influence a wind control decision result of a server, the client can be prevented from purposefully modifying original data to cause pollution of a server database and disordered model training, on the other hand, network bandwidth, system calculation and storage pressure can be greatly reduced through pre-clearing of bad bugs, and user experience of real-time application is improved.
Referring to fig. 7 in conjunction with the foregoing embodiments, fig. 7 is a schematic flowchart of a wind control decision method according to another embodiment of the present application. The wind control decision method can be applied to a client. As shown in FIG. 2, the client and server are communicatively coupled. The client is configured with a corresponding browser and/or APP, the browser and/or APP running can send a service request to the server, and the server process can respond to the service request.
Illustratively, the business request includes at least one of internet financial services such as a network credit application, a payment application, a financing application, and a financial insurance purchase. For example, the user's application of Html5 at the client triggers the client to send a service request to the server.
The H5 application can support the access of the foot bugs (bots), can fully exert the propagation characteristics of the Internet and has strong information popularization efficiency. As a good wind control method and application, supporting H5 will become an important market demand. Native APP can be avoided because by native program parcel, form the data isolated island, lack the agreeable nature with the internet, the not enough of visibility is low.
For example, the wind control decision method of the present embodiment is mainly applied to a scenario where the client runs the H5 application.
As shown in fig. 7, the wind control decision method specifically includes steps S210 to S250.
Step S210, determining whether the current user is a bad machine user, and rejecting the bad machine user.
Step S220, obtaining original data of the current user, where the original data includes at least one of active data, interactive data, and passive data.
The active data comprise data submitted by a user, the interactive data comprise user operation actions, and the passive data comprise client information.
And step S230, verifying the active data and sending the verified active data to a server.
Step S240, performing a structuring process on the interactive data and/or the passive data to obtain structured data.
Step S250, based on a preset uploading rule, sending the structured data to the server, so that the server verifies the active data and analyzes the structured data, based on a portrait model in a portrait model library, generating user characteristic data according to the verified active data and the analyzed data to obtain a user portrait of the current user, and based on a wind control decision engine, the server generates a wind control result according to the user portrait and responds to a service request of the client according to the wind control result.
In some possible embodiments, the structuring the interactive data and/or the passive data includes: and at least one of classification, packaging, compression, temporary storage and combination is carried out on the interactive data and/or the passive data.
Illustratively, the preset upload rule includes: and sending the structured data to the server according to the data volume of the structured data, the last time of sending the structured data and/or whether a key event is detected.
The specific principle and implementation manner of the wind control decision method for the client provided in the embodiment of the present specification are all matched with or similar to the wind control decision method for the server in the foregoing embodiment, and are not described herein again.
The methods of the present application are operational with numerous general purpose or special purpose computing system environments or configurations. For example: personal computers, server computers, hand-held or portable devices, tablet-type devices, multiprocessor systems, microprocessor-based systems, set top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments that include any of the above systems or devices, and the like.
Illustratively, the above-described method may be implemented in the form of a computer program that is executable on a computer device as shown in fig. 8.
Referring to fig. 8, fig. 8 is a schematic structural diagram of a computer device according to an embodiment of the present disclosure. The computer device may be a server or a terminal.
Referring to fig. 8, the computer device includes a processor, a memory, and a network interface connected through a system bus, wherein the memory may include a nonvolatile storage medium and an internal memory.
The non-volatile storage medium may store an operating system and a computer program. The computer program includes program instructions that, when executed, cause a processor to perform any of the wind control decision methods.
The processor is used for providing calculation and control capability and supporting the operation of the whole computer equipment.
The internal memory provides an environment for the execution of a computer program on a non-volatile storage medium, which when executed by the processor causes the processor to perform any of the methods of the wind-controlled decision-making.
The network interface is used for network communication, such as sending assigned tasks and the like. Those skilled in the art will appreciate that the configuration of the computer apparatus is merely a block diagram of a portion of the configuration associated with aspects of the present application and is not intended to limit the computer apparatus to which aspects of the present application may be applied, and that a particular computer apparatus may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
It should be understood that the Processor may be a Central Processing Unit (CPU), and the Processor may be other general purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components, etc. Wherein a general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
Wherein, in one embodiment, the processor is configured to execute a computer program stored in the memory to implement the steps of:
receiving active data and structured data from a client, wherein the active data comprises data submitted by a user, the structured data comprises data obtained by the client performing structured processing on interactive data and/or passive data, the interactive data comprises user operation actions, and the passive data comprises client information;
verifying the active data, analyzing the structured data, and storing the verified active data and the analyzed data obtained by analysis in a database;
generating user characteristic data according to the verified active data and the analyzed data based on the portrait model in the portrait model library so as to establish a user portrait of the current user;
generating a wind control result according to the user portrait based on a wind control decision engine;
and responding the service request of the client according to the wind control result.
Illustratively, the wind control decision engine comprises at least one decision tree comprising at least one node;
the wind control decision engine is used for generating a wind control result according to the user portrait, and comprises the following steps:
generating a corresponding wind control variable according to corresponding user characteristic data based on a decision node of the decision tree, and adjusting a wind control coefficient of the decision tree according to the wind control variable;
and generating a wind control result according to the wind control coefficient of at least one decision tree.
Wherein, in one embodiment, the processor is configured to execute a computer program stored in the memory to implement the steps of:
determining whether the current user is a bad machine user, and rejecting the bad machine user;
the method comprises the steps of obtaining original data of a current user, wherein the original data comprise at least one of active data, interactive data and passive data, the active data comprise data submitted by the user, the interactive data comprise user operation actions, and the passive data comprise client information;
verifying the active data and sending the verified active data to a server;
carrying out structuring processing on the interactive data and/or the passive data to obtain structured data;
the method comprises the steps of sending structured data to a server based on a preset uploading rule, verifying the active data and analyzing the structured data by the server, generating user feature data according to the verified active data and the analyzed data to obtain a user portrait of a current user based on a portrait model in a portrait model library, generating a wind control result according to the user portrait by the server based on a wind control decision engine, and responding to a service request of a client according to the wind control result.
From the above description of the embodiments, it is clear to those skilled in the art that the present application can be implemented by software plus necessary general hardware platform. Based on such understanding, the technical solution of the present application may be substantially implemented or contributed to by the prior art in the form of a software product, which may be stored in a storage medium, such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method according to the embodiments or some parts of the embodiments of the present application, such as:
a computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and the computer program includes program instructions, and the processor executes the program instructions to implement any of the wind control decision methods provided in the embodiments of the present application.
The computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiment, for example, a hard disk or a memory of the computer device. The computer readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like provided on the computer device.
While the invention has been described with reference to specific embodiments, the scope of the invention is not limited thereto, and those skilled in the art can easily conceive various equivalent modifications or substitutions within the technical scope of the invention. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (10)
1. A method for a wind control decision, for a server, the method comprising:
receiving active data and structured data from a client, wherein the active data comprises data submitted by a user, the structured data comprises data obtained by the client performing structured processing on interactive data and/or passive data, the interactive data comprises user operation actions, and the passive data comprises client information;
verifying the active data, analyzing the structured data, and storing the verified active data and the analyzed data obtained by analysis in a database;
generating user characteristic data according to the verified active data and the analyzed data based on the portrait model in the portrait model library so as to establish a user portrait of the current user;
generating a wind control result according to the user portrait based on a wind control decision engine;
responding the service request of the client according to the wind control result;
wherein the wind control decision engine comprises at least one decision tree comprising at least one node;
the wind control decision engine is used for generating a wind control result according to the user portrait, and comprises the following steps:
generating a corresponding wind control variable according to corresponding user characteristic data based on a decision node of the decision tree, and adjusting a wind control coefficient of the decision tree according to the wind control variable;
and generating a wind control result according to the wind control coefficient of at least one decision tree.
2. The wind control decision method of claim 1, wherein the generating a wind control result from the user representation based on a wind control decision engine further comprises:
based on the decision node of the decision tree, skipping to the next decision node according to the corresponding user characteristic data; or
And generating a wind control result according to the corresponding user characteristic data based on the decision nodes of the decision tree.
3. The wind control decision method according to claim 1, wherein the decision nodes comprise rule nodes and/or score nodes, and the generating of the corresponding wind control variables from the corresponding user characteristic data based on the decision nodes of the decision tree comprises:
generating corresponding wind control variables according to corresponding user characteristic data based on the rule nodes and/or the scoring nodes;
the rule node comprises a plurality of wind control rules and combinational logic of the plurality of wind control rules, and the scoring node comprises a plurality of scoring factors and respective weights of the scoring factors and the combinational logic of the scoring factors.
4. The wind control decision method of claim 3, further comprising:
determining a plurality of wind control rules and the combinational logic of the plurality of wind control rules from a rule base based on a machine learning algorithm and/or the management operation of wind control personnel to generate new rule nodes; and/or
And determining a plurality of scoring factors, the weights of the scoring factors and the combination logic of the scoring factors from a scoring factor library based on the management operation of a machine learning algorithm and/or a wind control person to generate a new scoring node.
5. The wind control decision method of any one of claims 1-4, further comprising:
acquiring third-party data and/or historical data of the current user;
and generating user characteristic data according to the analysis data and the third party data and/or the historical data based on the portrait model in the portrait model base so as to establish the user portrait of the current user.
6. The wind control decision method of any one of claims 1-4, further comprising:
deleting, updating or adding the portrait model in the portrait model library;
wherein, the deleting, updating or adding the portrait model in the portrait model library comprises:
training a portrait model according to portrait labels of a plurality of users and active data and/or the analytic data based on a machine learning algorithm;
evaluating the effect of the portrait model, and updating or adding the portrait model which is evaluated to the portrait model library.
7. A method for a wind control decision-making, for a client, the method comprising:
determining whether the current user is a bad machine user, and rejecting the bad machine user;
the method comprises the steps of obtaining original data of a current user, wherein the original data comprise at least one of active data, interactive data and passive data, the active data comprise data submitted by the user, the interactive data comprise user operation actions, and the passive data comprise client information;
verifying the active data and sending the verified active data to a server;
carrying out structuring processing on the interactive data and/or the passive data to obtain structured data;
the method comprises the steps of sending structured data to a server based on a preset uploading rule, verifying the active data and analyzing the structured data by the server, generating user feature data according to the verified active data and the analyzed data to obtain a user portrait of a current user based on a portrait model in a portrait model library, generating a wind control result according to the user portrait by the server based on a wind control decision engine, and responding to a service request of a client according to the wind control result.
8. A computer device, wherein the computer device comprises a memory and a processor;
the memory is used for storing a computer program;
the processor for executing the computer program and for implementing the wind control decision method according to any of claims 1-6 when executing the computer program.
9. A computer device, wherein the computer device comprises a memory and a processor;
the memory is used for storing a computer program;
the processor for executing the computer program and when executing the computer program implementing the wind control decision method of claim 7.
10. A computer-readable storage medium storing a computer program, wherein if the computer program is executed by a processor, the computer program implements:
the wind control decision method of any one of claims 1-6; and/or
The wind control decision method of claim 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010405458.0A CN111311136A (en) | 2020-05-14 | 2020-05-14 | Wind control decision method, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010405458.0A CN111311136A (en) | 2020-05-14 | 2020-05-14 | Wind control decision method, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111311136A true CN111311136A (en) | 2020-06-19 |
Family
ID=71159394
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010405458.0A Pending CN111311136A (en) | 2020-05-14 | 2020-05-14 | Wind control decision method, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111311136A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111738812A (en) * | 2020-08-21 | 2020-10-02 | 深圳索信达数据技术有限公司 | Information pushing method and system based on user group micro-segmentation |
| CN111813399A (en) * | 2020-07-23 | 2020-10-23 | 平安医疗健康管理股份有限公司 | Machine learning-based auditing rule processing method and device and computer equipment |
| CN111861701A (en) * | 2020-07-09 | 2020-10-30 | 深圳市富之富信息技术有限公司 | Wind control model optimization method and device, computer equipment and storage medium |
| CN112348403A (en) * | 2020-11-26 | 2021-02-09 | 德联易控科技(北京)有限公司 | Wind control model construction method and device and electronic equipment |
| CN112380167A (en) * | 2020-11-17 | 2021-02-19 | 深圳市和讯华谷信息技术有限公司 | Batch data verification method and device, computer equipment and storage medium |
| CN112446613A (en) * | 2020-11-26 | 2021-03-05 | 深圳华锐金融技术股份有限公司 | External access client wind control method, device, equipment and storage medium |
| CN112907347A (en) * | 2020-12-02 | 2021-06-04 | 浙江惠瀜网络科技有限公司 | Data management system for client portrait of automobile installments and data processing method based on system |
| CN113362137A (en) * | 2021-06-11 | 2021-09-07 | 北京十一贝科技有限公司 | Insurance product recommendation method and device, terminal equipment and storage medium |
| CN113657779A (en) * | 2021-08-20 | 2021-11-16 | 杭州时趣信息技术有限公司 | Dynamically-configured wind control decision method, system, equipment and storage medium |
| CN114095282A (en) * | 2022-01-21 | 2022-02-25 | 杭银消费金融股份有限公司 | Wind control processing method and device based on short text feature extraction |
| CN114884964A (en) * | 2022-07-11 | 2022-08-09 | 上海富友支付服务股份有限公司 | Service wind control method and system based on Tuxedo architecture |
| CN116579838A (en) * | 2023-05-17 | 2023-08-11 | 深圳市三希软件科技有限公司 | Intelligent supervision method and system based on big data |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010097450A (en) * | 2008-10-17 | 2010-04-30 | Intelligent Wave Inc | Settlement approval system for credit card, credit card to be used for settlement approval system for credit card, terminal apparatus, host computer, and settlement approval method for credit card |
| CN106952053A (en) * | 2017-04-07 | 2017-07-14 | 北京工业大学 | A kind of Supply chain financing method based on dynamic flow |
| CN107392755A (en) * | 2017-07-07 | 2017-11-24 | 南京甄视智能科技有限公司 | Credit risk merges appraisal procedure and system |
| CN107819727A (en) * | 2016-09-13 | 2018-03-20 | 腾讯科技（深圳）有限公司 | A kind of network safety protection method and system based on the safe credit worthiness of IP address |
| CN108596434A (en) * | 2018-03-23 | 2018-09-28 | 卫盈联信息技术（深圳）有限公司 | Fraud detection and methods of risk assessment, system, equipment and storage medium |
| CN109857373A (en) * | 2018-12-18 | 2019-06-07 | 平安科技（深圳）有限公司 | Business data processing method, device, computer equipment and storage medium |
| CN110008991A (en) * | 2019-02-26 | 2019-07-12 | 阿里巴巴集团控股有限公司 | The identification of risk case, risk identification model generation method and device |
| CN110399925A (en) * | 2019-07-26 | 2019-11-01 | 腾讯科技（武汉）有限公司 | Risk Identification Method, device and the storage medium of account |
| CN110648217A (en) * | 2019-09-08 | 2020-01-03 | 贵州普惠链电子商务有限公司 | Wind control system based on big data and artificial intelligence |
-
2020
- 2020-05-14 CN CN202010405458.0A patent/CN111311136A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010097450A (en) * | 2008-10-17 | 2010-04-30 | Intelligent Wave Inc | Settlement approval system for credit card, credit card to be used for settlement approval system for credit card, terminal apparatus, host computer, and settlement approval method for credit card |
| CN107819727A (en) * | 2016-09-13 | 2018-03-20 | 腾讯科技（深圳）有限公司 | A kind of network safety protection method and system based on the safe credit worthiness of IP address |
| CN106952053A (en) * | 2017-04-07 | 2017-07-14 | 北京工业大学 | A kind of Supply chain financing method based on dynamic flow |
| CN107392755A (en) * | 2017-07-07 | 2017-11-24 | 南京甄视智能科技有限公司 | Credit risk merges appraisal procedure and system |
| CN108596434A (en) * | 2018-03-23 | 2018-09-28 | 卫盈联信息技术（深圳）有限公司 | Fraud detection and methods of risk assessment, system, equipment and storage medium |
| CN109857373A (en) * | 2018-12-18 | 2019-06-07 | 平安科技（深圳）有限公司 | Business data processing method, device, computer equipment and storage medium |
| CN110008991A (en) * | 2019-02-26 | 2019-07-12 | 阿里巴巴集团控股有限公司 | The identification of risk case, risk identification model generation method and device |
| CN110399925A (en) * | 2019-07-26 | 2019-11-01 | 腾讯科技（武汉）有限公司 | Risk Identification Method, device and the storage medium of account |
| CN110648217A (en) * | 2019-09-08 | 2020-01-03 | 贵州普惠链电子商务有限公司 | Wind control system based on big data and artificial intelligence |
Non-Patent Citations (3)
| Title |
|---|
| PXHOONET: "表单验证放在服务器端有什么好处？", 《百度知道-HTTPS://ZHIDAO.BAIDU.COM/QUESTION/80946095.HTML》 * |
| 嘶吼ROARTALK: "浅谈风控领域中的用户画像", 《安全脉搏-HTTPS://WWW.SECPULSE.COM/ARCHIVES/51979.HTML》 * |
| 晓冉1008: "电脑成了肉鸡的一些症状是什么。可以用杀毒软件检测出来吗？", 《百度知道-HTTPS://ZHIDAO.BAIDU.COM/QUESTION/552208716525556052.HTML》 * |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111861701A (en) * | 2020-07-09 | 2020-10-30 | 深圳市富之富信息技术有限公司 | Wind control model optimization method and device, computer equipment and storage medium |
| CN111813399A (en) * | 2020-07-23 | 2020-10-23 | 平安医疗健康管理股份有限公司 | Machine learning-based auditing rule processing method and device and computer equipment |
| CN111813399B (en) * | 2020-07-23 | 2022-05-31 | 平安医疗健康管理股份有限公司 | Machine learning-based auditing rule processing method and device and computer equipment |
| CN111738812A (en) * | 2020-08-21 | 2020-10-02 | 深圳索信达数据技术有限公司 | Information pushing method and system based on user group micro-segmentation |
| CN112380167A (en) * | 2020-11-17 | 2021-02-19 | 深圳市和讯华谷信息技术有限公司 | Batch data verification method and device, computer equipment and storage medium |
| CN112380167B (en) * | 2020-11-17 | 2024-06-25 | 深圳市和讯华谷信息技术有限公司 | Batch data verification method and device, computer equipment and storage medium |
| CN112348403A (en) * | 2020-11-26 | 2021-02-09 | 德联易控科技(北京)有限公司 | Wind control model construction method and device and electronic equipment |
| CN112446613A (en) * | 2020-11-26 | 2021-03-05 | 深圳华锐金融技术股份有限公司 | External access client wind control method, device, equipment and storage medium |
| CN112907347B (en) * | 2020-12-02 | 2023-07-04 | 浙江惠瀜网络科技有限公司 | System for managing image data of client of automobile staged loan and data processing method based on system |
| CN112907347A (en) * | 2020-12-02 | 2021-06-04 | 浙江惠瀜网络科技有限公司 | Data management system for client portrait of automobile installments and data processing method based on system |
| CN113362137A (en) * | 2021-06-11 | 2021-09-07 | 北京十一贝科技有限公司 | Insurance product recommendation method and device, terminal equipment and storage medium |
| CN113362137B (en) * | 2021-06-11 | 2024-04-05 | 北京十一贝科技有限公司 | Insurance product recommendation method and device, terminal equipment and storage medium |
| CN113657779A (en) * | 2021-08-20 | 2021-11-16 | 杭州时趣信息技术有限公司 | Dynamically-configured wind control decision method, system, equipment and storage medium |
| CN113657779B (en) * | 2021-08-20 | 2024-01-09 | 杭州时趣信息技术有限公司 | Dynamic configuration wind control decision method, system, equipment and storage medium |
| CN114095282B (en) * | 2022-01-21 | 2022-04-15 | 杭银消费金融股份有限公司 | Wind control processing method and device based on short text feature extraction |
| CN114095282A (en) * | 2022-01-21 | 2022-02-25 | 杭银消费金融股份有限公司 | Wind control processing method and device based on short text feature extraction |
| CN114884964A (en) * | 2022-07-11 | 2022-08-09 | 上海富友支付服务股份有限公司 | Service wind control method and system based on Tuxedo architecture |
| CN116579838A (en) * | 2023-05-17 | 2023-08-11 | 深圳市三希软件科技有限公司 | Intelligent supervision method and system based on big data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111311136A (en) | Wind control decision method, computer equipment and storage medium | |
| CN110399925B (en) | Account risk identification method, device and storage medium | |
| CN107818344B (en) | Method and system for classifying and predicting user behaviors | |
| CN111681091B (en) | Financial risk prediction method and device based on time domain information and storage medium | |
| CN110442712B (en) | Risk determination method, risk determination device, server and text examination system | |
| US11574274B2 (en) | System and method for evaluating application errors in e-commerce applications | |
| CN113590497A (en) | Business service test method and device, electronic equipment and storage medium | |
| CN108334641B (en) | Method, system, electronic equipment and storage medium for collecting user behavior data | |
| CN103685307A (en) | Method, system, client and server for detecting phishing fraud webpage based on feature library | |
| CN109831459B (en) | Method, device, storage medium and terminal equipment for secure access | |
| CN111522724B (en) | Method and device for determining abnormal account number, server and storage medium | |
| CN111666298A (en) | Method and device for detecting user service class based on flink, and computer equipment | |
| CN111581258B (en) | Security data analysis method, device, system, equipment and storage medium | |
| CN115982012A (en) | Evaluation model and method for interface management capability maturity | |
| CN113918794B (en) | Enterprise network public opinion benefit analysis method, system, electronic equipment and storage medium | |
| CN107038377B (en) | Website authentication method and device and website credit granting method and device | |
| CN114422271A (en) | Data processing method, device, equipment and readable storage medium | |
| CN112347457A (en) | Abnormal account detection method and device, computer equipment and storage medium | |
| CN113850669A (en) | User grouping method and device, computer equipment and computer readable storage medium | |
| CN116318974A (en) | Site risk identification method and device, computer readable medium and electronic equipment | |
| CN116228391A (en) | Risk identification method and device, storage medium and electronic equipment | |
| CN115689571A (en) | Abnormal user behavior monitoring method, device, equipment and medium | |
| CN111522747B (en) | Application processing method, device, equipment and medium | |
| CN112199573B (en) | Illegal transaction active detection method and system | |
| CN114493853A (en) | Credit rating evaluation method, credit rating evaluation device, electronic device and storage medium |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200619 |