WO2021218312A1

WO2021218312A1 - Method and apparatus for constructing service fraud identification database, and computer device

Info

Publication number: WO2021218312A1
Application number: PCT/CN2021/077419
Authority: WO
Inventors: 黄丽芝
Original assignee: 深圳壹账通智能科技有限公司
Priority date: 2020-04-27
Filing date: 2021-02-23
Publication date: 2021-11-04
Also published as: CN111552680A

Abstract

The present application relates to the technical field of data processing, and can be applied to a smart security protection scenario. Provided is a method for constructing a service fraud identification database. The method comprises: collecting service fraud event information, wherein the service fraud event information carries a user identifier; acquiring a key information identifier of service fraud information, and according to the key information identifier, extracting corresponding key information from the service fraud event information; performing feature extraction on the key information to obtain feature information of the service fraud event information; acquiring a feature code of the feature information, and determining a feature label of the service fraud event information according to the feature code of the feature information; querying the correlation between a preset feature label and a service fraud behavior label, and determining a service fraud behavior label of the service fraud event information; and constructing a service fraud identification database according to the user identifier and the service fraud behavior label. The present method is based on data processing technology, and improves the efficiency of identifying a service fraud behavior.

Description

Method, device and computer equipment for constructing business fraud identification database

This application claims the priority of a Chinese patent application filed with the Chinese Patent Office on April 27, 2020, the application number is 202010344725.8, and the invention title is "Method, Apparatus, and Computer Equipment for Building a Business Fraud Recognition Database", the entire content of which is incorporated by reference Incorporated in this application.

Technical field

This application relates to the field of data processing technology, and in particular to a method, device, computer equipment and storage medium for constructing a business fraud identification database.

Background technique

With the development of society, there are more and more people handling business (such as insurance), and business fraud is also increasing. In order to identify business fraud, a database for identifying business fraud can be constructed.

The inventor found that, currently, the method of constructing a database for identifying business frauds is generally to obtain a large amount of manually collected data through a server, such as users' online information and offline survey results, etc., and perform calculation and analysis on the large amount of data. In order to obtain the user's business fraud identification result, and build a database for identifying business fraud according to the user's business fraud identification result. However, the inventor realizes that the resources of the server are limited. If in the process of constructing the database, it is necessary to analyze and identify a large amount of manually collected data for each user, which will consume a lot of time, resulting in the efficiency of identifying business fraud. Lower.

technical problem

Based on this, it is necessary to provide a method, device, computer equipment, and storage medium for constructing a business fraud identification database that can improve the efficiency of business fraud identification in response to the above technical problems.

Technical solutions

A method for constructing a business fraud identification database, the method comprising: collecting business fraud event information; the business fraud event information carries a user identification; acquiring the key information identifier of the business fraud event information, and according to the key The information identifier extracts the corresponding key information from the business fraud event information; extracts the features of the key information to obtain the feature information of the business fraud event information; obtains the feature code of the feature information, according to the The feature code of the feature information is used to determine the feature tag of the business fraud event information; according to the feature tag of the business fraud event information, the correspondence between the preset feature tag and the business fraud behavior tag is inquired to determine the business fraud event The business fraud behavior label of the information; according to the user identification of the business fraud event information and the business fraud behavior label, a business fraud identification database is constructed.

A device for constructing a business fraud identification database. The device includes: an information collection module for collecting business fraud event information; the business fraud event information carries user identification; an information extraction module for acquiring the business fraud The key information identifier of the event information, according to the key information identifier, extracts the corresponding key information from the business fraud event information; the information determination module is used for feature extraction of the key information to obtain the business Characteristic information of fraud event information; a characteristic label determining module for acquiring the characteristic code of the characteristic information, and determining the characteristic label of the business fraud event information according to the characteristic code of the characteristic information; a behavior label determining module for According to the feature tag of the business fraud event information, query the correspondence between the preset feature tag and the business fraud behavior tag, and determine the business fraud behavior tag of the business fraud event information; the database building module is used to determine the business fraud behavior tag of the business fraud event information; The user identification and the business fraud behavior label of the event information construct a business fraud identification database.

A computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the following steps when executing the computer program: collecting business fraud event information; the business fraud event information carries a user identification Obtain the key information identifier of the business fraud event information, and extract the corresponding key information from the business fraud event information according to the key information identifier; perform feature extraction on the key information to obtain the business The characteristic information of the fraud event information; obtain the characteristic code of the characteristic information, and determine the characteristic label of the business fraud event information according to the characteristic code of the characteristic information; query the preset according to the characteristic label of the business fraud event information The corresponding relationship between the feature label and the business fraud label to determine the business fraud label of the business fraud event information; construct a business fraud identification database according to the user identification of the business fraud event information and the business fraud label .

A computer-readable storage medium, on which a computer program is stored, when the computer program is executed by a processor, the following steps are implemented: collecting business fraud event information; the business fraud event information carries a user identification; acquiring the business The key information identifier of the fraud event information, according to the key information identifier, extract the corresponding key information from the business fraud event information; perform feature extraction on the key information to obtain the feature of the business fraud event information Information; obtain the feature code of the feature information, determine the feature label of the business fraud event information according to the feature code of the feature information; query the preset feature label and business according to the feature label of the business fraud event information Correspondence of the fraudulent behavior label, determine the business fraud behavior label of the business fraud event information; construct a business fraud identification database according to the user identification of the business fraud event information and the business fraud behavior label.

Beneficial effect

This application does not need to analyze and identify a large amount of manually collected data, simplifies the business fraud identification process, and further improves the identification efficiency of business fraud.

Description of the drawings

Fig. 1 is a schematic flowchart of a method for constructing a business fraud identification database in an embodiment.

Fig. 2 is a schematic flowchart of the steps of sending a business fraud identification result to a terminal in an embodiment.

Fig. 3 is a schematic flowchart of a method for identifying business fraud in an embodiment.

Fig. 4 is a structural block diagram of an apparatus for constructing a business fraud identification database in an embodiment.

Fig. 5 is an internal structure diagram of a computer device in an embodiment.

Embodiments of the present invention

In order to make the purpose, technical solutions, and advantages of this application clearer and clearer, the following further describes the application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, and are not used to limit the present application.

The technical solution of this application relates to the field of artificial intelligence and/or big data technology, and can be applied to scenarios such as smart security such as information monitoring to promote the construction of smart cities. Optionally, the data involved in this application, such as business fraud event information, feature tags, and/or behavior tags, can be stored in a database, or can be stored in a blockchain, such as distributed storage through a blockchain, this application does not Make a limit.

In one embodiment, as shown in FIG. 1, a method for constructing a business fraud identification database is provided. This embodiment uses the method applied to a server for illustration. It is understandable that the method can also be applied to a terminal. It can also be applied to a system including a terminal and a server, and is realized through the interaction between the terminal and the server. In this embodiment, the method includes the following steps.

Step S101: Collect business fraud event information; the business fraud event information carries a user identifier.

In this step, the business fraud event information refers to the event information of financial business fraud, such as loan fraud event information, insurance fraud event information, etc., specifically it can be intentional risk event information, repeated insurance coverage event information, forged evidence event information, Repeat claim event information, insurance backlog event information, loan overdue event information, etc. User identification refers to user information that is used to identify business fraud event information, such as name, ID number, etc.

Specifically, the server collects business event information on the network based on big data technology; identifies the information type of the business event information, and filters out the business event information whose information type matches the information type of the business fraud event information from the collected business event information , As business fraud event information; for example, filter out business event information whose information type matches insurance fraud event information, and use it as business fraud event information; identify the business fraud event information to obtain the user ID corresponding to the business fraud event information.

Further, the server can also obtain preset data preprocessing instructions, and perform data preprocessing on the collected business fraud event information according to the preset data preprocessing instructions, such as data cleaning, data standardization processing, etc., after data preprocessing is obtained Information about business fraud incidents.

Step S102: Obtain the key information identifier of the business fraud event information, and extract the corresponding key information from the business fraud event information according to the key information identifier.

In this step, the key information refers to the business information in the business fraud event information. For example, the business fraud event information is insurance fraud event information, and the key information is insurance business information in the insurance fraud event information. The key information identifier is used to identify the key information in the business fraud event information, and is an identifier that can automatically identify the key information in the business fraud event information.

Specifically, the server obtains the key information identifier of the business fraud event information in advance, and extracts information corresponding to the key information identifier from the business fraud event information as the key information in the business fraud event information. In this way, it is beneficial to extract the key information in the business fraud event information, avoid unnecessary information interference, and further simplify the business fraud identification process, thereby improving the identification efficiency of business fraud.

Step S103: Perform feature extraction on key information to obtain feature information of business fraud event information.

In this step, the characteristic information refers to the user's fraud behavior characteristic information corresponding to the business fraud event information. Different key information has different characteristic information corresponding to the business fraud event information. Let their relatives, friends or other relevant personnel pretend to be information on the behavioral characteristics of medical treatment, etc.

Specifically, the server obtains a preset feature information extraction instruction, performs feature extraction on key information according to the preset feature information extraction instruction, and obtains feature information of business fraud event information; wherein, the preset feature information extraction instruction is a kind of Capable of extracting instructions for characteristic information corresponding to key information. In this way, by determining the characteristic information of the business fraud event information, it is beneficial to subsequently determine the characteristic label of the business fraud event information based on the characteristic information of the business fraud event information, and then determine the business fraud behavior label of the user corresponding to the business fraud event information.

In an embodiment, the server may also input key information into a pre-trained feature extraction network (such as a convolutional neural network), and perform feature extraction on the key information through the pre-trained feature extraction network to obtain feature information of business fraud event information.

In one embodiment, the server can also identify the characteristic information in the key information based on the information recognition technology, and use the characteristic information as the characteristic information of the business fraud event information; it is convenient to subsequently determine the business fraud based on the characteristic information of the business fraud event information The feature label of the event information.

Further, the server can also obtain the correspondence between the preset key information and the characteristic information; query the correspondence between the preset key information and the characteristic information, determine the characteristic information corresponding to the key information, and use it as the characteristic information of the business fraud event information; The server pre-builds the correspondence between the key information and characteristic information of the business fraud event information as the correspondence between the preset key information and the characteristic information; queries the preset key information and characteristic information according to the key information of the business fraud event information Correspondence, determine the characteristic information of business fraud event information.

Step S104: Obtain the feature code of the feature information, and determine the feature tag of the business fraud event information according to the feature code of the feature information.

In this step, the feature encoding of feature information refers to a low-dimensional feature vector used to represent low-level semantics of feature information after compression encoding, which can be learned through a pre-trained feature embedding network model. The feature tag is used to identify the user's fraudulent behavior feature information corresponding to the business fraud event information, such as deliberate insurance, repeated insurance, forged evidence, repeated claims, insurance backlog, loan overdue, etc.

Specifically, the server obtains the preset feature encoding instruction, performs encoding processing on the feature information according to the preset feature encoding instruction, and obtains the feature code of the feature information; performs full connection processing on the feature code of the feature information, and obtains that the feature information belongs to each preset. Set the probability of the feature tag, and use the preset feature tag with the highest probability as the feature tag of the business fraud event information. Among them, the preset feature encoding instruction is an instruction capable of encoding information.

Further, in the process of determining the feature label of the business fraud event information, the server can also obtain the feature information of the preset feature tag, and match the feature information of the business fraud event information with the feature information of the preset feature tag. If the feature information of the event information is successfully matched with the feature information of the preset feature tag, the preset feature tag is identified as the feature tag of the business fraud event information.

For example, if the feature information of the business fraud event information matches the feature information corresponding to the preset feature tag "repetitive insurance", then the "duplicate insurance" is identified as the feature tag of the business fraud event information.

Step S105, according to the characteristic label of the business fraud event information, query the correspondence between the preset characteristic label and the business fraud behavior label, and determine the business fraud behavior label of the business fraud event information.

In this step, the business fraud label is used to identify the user’s business fraud, which can be a fraud risk label, a malicious behavior label, a credit risk label, an external list label, etc.; different feature labels have different corresponding business fraud labels ; For example, inverted bills, group fraudulent loans, expanded losses, and undetermined loss prosecution are fraud risk labels; malicious house purchases, malicious house purchases, driving without a license are malicious behavior labels; loan defaults and long-term loan defaults are credit risk labels; Ministry of Public Security The list of dangerous elements and the persons involved in economic cases belong to external list labels.

Specifically, the server obtains the correspondence between the preset feature label and the business fraud activity label, and according to the feature label of the business fraud event information, queries the correspondence between the preset feature label and the business fraud activity label, and determines the corresponding relationship with the characteristic label. The business fraud label is used as the business fraud label of the business fraud event information.

Further, before querying the correspondence between the preset feature tags and the business fraud behavior tags according to the feature tags of the business fraud event information, and determining the business fraud behavior tags of the business fraud event information, the server may also obtain multiple feature tags and each According to the business fraud labels marked by the feature labels, the business fraud labels corresponding to the multiple feature labels are obtained according to the business fraud labels marked by the feature labels; the feature labels are constructed according to the business fraud labels corresponding to the multiple feature labels. The correspondence relationship between the business fraud behavior labels is used as the correspondence relationship between the preset feature labels and the business fraud behavior labels.

Step S106: Construct a business fraud identification database according to the user identification and business fraud behavior label of the business fraud event information.

Specifically, the server categorizes and stores business fraud labels in the business fraud identification database according to the user identification of the business fraud event information, so as to store business fraud labels corresponding to multiple user identifications in the business fraud identification database, so as to facilitate the identification of users according to By querying the business fraud identification database with the user ID, you can determine whether the business fraud identification database stores the business fraud label corresponding to the user ID, and then determine whether the user to be identified has business fraud, which simplifies the business fraud identification process and further improves Improve the efficiency of business fraud identification.

Further, after a certain period of time, such as 2 months, the server re-collects the business fraud event information on the network, and re-executes the above steps S102 to S105 to obtain the business fraud behavior label and user identification of the business fraud event information, according to the business fraud The business fraud label and user ID of the event information, and the update of the business fraud identification database is conducive to improving the timeliness and accuracy of the business fraud label of the user ID stored in the business fraud identification database, and further improves the subsequent business fraud identification The accuracy of the results.

In the above-mentioned method for constructing a business fraud identification database, the characteristic label of the business fraud event information is determined through the key information in the collected business fraud event information, and then the business fraud behavior label of the business fraud event information is determined; according to the user of the business fraud event information Identify and business fraud identification database to construct a business fraud identification database; in this way, the business fraud identification database constructed can store business fraud identification tags corresponding to multiple user identifications, which is convenient for querying the business fraud identification database based on the user identification of the user to be identified. It can be determined whether the business fraud identification database stores the business fraud label corresponding to the user ID, and then determine whether the user to be identified has business fraud; there is no need to analyze and identify a large amount of manually collected data, which simplifies the business fraud identification process. Further improve the identification efficiency of business fraud. The proposed solution can be applied to scenarios such as smart security, such as information monitoring, so as to promote the construction of smart cities.

In one embodiment, the above step S101, collecting business fraud event information, includes: collecting multiple business event information; identifying the information content of the business event information, and determining the information type of the business event information according to the information content; The business event information whose information type matches the preset information type is screened out, and used as business fraud event information, and the business fraud event information is uploaded to the blockchain.

In this step, the preset information type is used to identify the information type of the business fraud event information, such as the intentional risk information type, the repeated insurance information type, the forged evidence information type, the repeated claim information type, the insurance reversed form information type, and the loan overdue information Type etc.

Specifically, the server collects business event information on the network based on big data technology; extracts the information content in the business event information, queries the preset correspondence relationship between the information content and the information type according to the information content, and determines the information type corresponding to the information content , As the information type of the business event information; match the information type of the business event information with the preset information type, if the matching is successful, the business event information is identified as business fraud event information; referring to this method, you can select from multiple The business fraud event information is filtered out from the business event information.

In this embodiment, collecting business fraud event information facilitates subsequent analysis of business fraud event information to determine the user's business fraud behavior label corresponding to the business fraud event information, without the need to analyze and identify a large amount of manually collected data , Thereby improving the efficiency of identifying business fraud.

In one embodiment, the above step S104, determining the feature label of the business fraud event information according to the feature code of the feature information, includes: inputting the feature code of the feature information into a pre-trained feature classification model to obtain the feature information in each preset feature The classification probability under the label; the preset feature label with the largest classification probability is used as the feature label of the business fraud event information.

Among them, the pre-trained feature classification model is a neural network model that can classify feature information to determine the feature label to which the feature information belongs. The classification probability of feature information under each preset feature label refers to the probability that the feature information belongs to each preset feature label.

Specifically, the server inputs the feature code of the feature information into the pre-trained feature classification model, and performs a series of neural network processing on the feature code of the feature information through the pre-trained feature classification model, such as convolution, pooling, full connection, etc., to obtain The classification code of the feature information under each preset feature label; normalize the classification code of the feature information under each preset feature label to obtain the classification probability of the feature information under each preset feature label; maximize the classification probability The preset feature tag of is used as the feature tag of business fraud event information.

In this embodiment, by determining the feature label of the business fraud event information, it is beneficial to subsequently determine the business fraud behavior label of the business fraud event information according to the feature label of the business fraud event information, and then construct a business fraud identification database. .

In one embodiment, the above step S106 is to construct a business fraud identification database based on the user identification of the business fraud event information and the business fraud behavior label, which includes: adding the business fraud behavior label of the business fraud event information to the user according to the business fraud event information The identification is stored in a preset database; the preset database is identified as a business fraud identification database.

In this embodiment, the business fraud identification database stores multiple user identification corresponding business fraud tags, which is convenient for querying the business fraud identification database based on the user identification of the user to be identified, so as to determine whether the user to be identified has business fraud. Improved the efficiency of identifying business fraud.

In one embodiment, as shown in FIG. 2, the method for constructing a business fraud identification database of the present application further includes the step of sending the business fraud identification result to the terminal, which specifically includes the following steps.

Step S201: Receive a service fraud identification request sent by a terminal; the service fraud identification request carries a user ID of the user to be identified.

In this step, the business fraud identification request is used to request the server to identify whether the user to be identified has business fraud.

Specifically, an inquiring user (such as a staff member of an insurance company) logs in to the terminal, and inputs the relevant user information identifier of the user to be identified (such as an insurance applicant) on the query operation interface of the terminal; the terminal recognizes the input operation of the inquiring user, and obtains the information to be identified The relevant user information identifier entered by the user is used as the user identifier; according to the user identifier, a business fraud identification request is generated, and the business fraud identification request is sent to the corresponding server. The server parses the received business fraud identification request to obtain the user identification of the user to be identified.

Step S202: Query the service fraud identification database according to the user ID to obtain the query result.

In this step, the query result is used to indicate whether a business fraud behavior label corresponding to the user ID is stored in the business fraud identification database.

Step S203: Determine the business fraud identification result of the user to be identified according to the query result.

In this step, the identification result of the business fraud of the user to be identified is used to indicate whether the user to be identified has business fraud, whether the user to be identified is a business fraud user, and so on.

Step S204: Send the identification result of business fraud to the terminal.

Specifically, the server sends the recognition result of the business fraud of the user to be identified to the terminal, and displays the recognition result of the business fraud of the user to be identified through the terminal; in this way, the querying user can quickly determine whether the user to be identified is a business fraud user without going through the line. Investigate or perform other operations.

In this embodiment, according to the business fraud identification request sent by the terminal, the business fraud identification database that stores the business fraud identification tags corresponding to multiple user identities can be queried, which can quickly determine whether the user to be identified is a business fraud user, and further improves the business. The efficiency of fraud identification.

In one embodiment, the above step S203, determining the business fraud identification result of the user to be identified based on the query result, includes: if the business fraud identification database is queried and stored in the business fraud identification database, the business fraud identification tag corresponding to the user identification is confirmed, Identify the user as a fraudulent user.

Specifically, if the query result is that the business fraud identification database stores a business fraud label corresponding to the user ID, it indicates that the user to be identified is a business fraud user, and there is a business fraud; if the query result is that there is no business fraud identification database with the user Identifies the corresponding business fraud label, indicating that the user to be identified is a normal user and there is no business fraud.

In this embodiment, the query result obtained by querying the business fraud identification database according to the user ID can quickly determine whether the user to be identified is a business fraud user, which further improves the identification efficiency of business fraud.

In one embodiment, after confirming that the user to be identified is a business fraud user, the method further includes: obtaining early warning information corresponding to the business fraud behavior label; then, the above step S204, sending the business fraud behavior identification result to the terminal, includes: The business fraud identification results and early warning information are sent to the terminal.

In this step, different business fraud labels have different warning information corresponding to them.

Specifically, after confirming that the user to be identified is a business fraud user, the server queries the correspondence between the preset business fraud label and the warning information, determines the warning information corresponding to the business fraud label identified by the user, and recognizes the business fraud The result is sent to the terminal together with the warning information, which is convenient to remind the querying user corresponding to the terminal; for example, the user is a high-risk fraud user, please be careful.

In this embodiment, after confirming that the user to be identified is a business fraud user, the corresponding early warning information is sent to the terminal, which is beneficial to promptly reminding the terminal corresponding inquiring user.

In one embodiment, as shown in FIG. 3, a method for identifying business fraud is provided, which includes the following steps.

Step S301: Collect multiple business event information; identify the information content of the business event information, and determine the information type of the business event information according to the information content; filter out the business event information whose information type matches the preset information type from the multiple business event information , As the business fraud event information, upload the business fraud event information to the blockchain.

Step S302: Obtain the key information identifier of the business fraud event information, and extract the corresponding key information from the business fraud event information according to the key information identifier.

Step S303: Perform feature extraction on key information to obtain feature information of business fraud event information.

Step S304: Obtain the feature code of the feature information, and determine the feature tag of the business fraud event information according to the feature code of the feature information.

Step S305, according to the feature tag of the business fraud event information, query the correspondence between the preset feature tag and the business fraud behavior tag, and determine the business fraud behavior tag of the business fraud event information.

Step S306: Store the business fraud behavior label of the business fraud event information in a preset database according to the user identification of the business fraud event information; identify the preset database as a business fraud identification database.

Step S307: Receive the service fraud identification request sent by the terminal; the service fraud identification request carries the user ID of the user to be identified.

Step S308, query the service fraud identification database according to the user ID to obtain the query result.

Step S309: According to the query result, determine the business fraud identification result of the user to be identified, and send the business fraud identification result to the terminal.

In this embodiment, according to the business fraud identification request sent by the terminal, the business fraud identification database that stores the business fraud identification tags corresponding to multiple user identities can be queried, which can quickly determine whether the user to be identified is a business fraud user, thereby simplifying the business. The fraud identification process saves a lot of time and further improves the efficiency of business fraud identification.

It should be understood that although the various steps in the flowcharts of FIGS. 1-3 are displayed in sequence as indicated by the arrows, these steps are not necessarily performed in sequence in the order indicated by the arrows. Unless specifically stated in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least some of the steps in Figures 1-3 can include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution of these steps or stages The sequence is not necessarily performed sequentially, but may be performed alternately or alternately with other steps or at least a part of the steps or stages in other steps.

In one embodiment, as shown in FIG. 4, a device for constructing a business fraud identification database is provided, which includes: an information collection module 410, an information extraction module 420, an information determination module 430, a feature tag identification module 440, and a behavior tag determination The module 450 and the database construction module 460, in which: the information collection module 410 is used to collect business fraud event information; the business fraud event information carries user identification; the information extraction module 420 is used to obtain the key information identification in the business fraud event information According to the key information identifier, the corresponding key information is extracted from the business fraud event information; the information determination module 430 is used for feature extraction of the key information to obtain the feature information of the business fraud event information; the feature label determination module 440, The feature code used to obtain the feature information, and the feature tag of the business fraud event information is determined according to the feature code of the feature information; the behavior tag determination module 450 is used to query the preset feature tag and business based on the feature tag of the business fraud event information Correspondence of fraudulent behavior tags to determine the business fraudulent behavior tags of the business fraud event information; the database construction module 460 is used to construct a business fraud identification database based on the user identification and the business fraudulent behavior tags of the business fraud event information.

In one embodiment, the information collection module 410 is also used to collect multiple business event information; identify the information content of the business event information, determine the information type of the business event information according to the information content; filter out the information type from the multiple business event information The business event information matching the preset information type is used as the business fraud event information, and the business fraud event information is uploaded to the blockchain.

In one embodiment, the feature label determination module 440 is further configured to input the feature code of the feature information into the pre-trained feature classification model to obtain the classification probability of the feature information under each preset feature label; and set the preset feature with the largest classification probability The label is used as the characteristic label of the business fraud event information.

In one embodiment, the database construction module 460 is further configured to store the business fraud behavior label of the business fraud event information in a preset database according to the user identification of the business fraud event information; and identify the preset database as a business fraud identification database.

In one embodiment, the device for constructing a business fraud identification database further includes a result sending module for receiving a business fraud identification request sent by the terminal; the business fraud identification request carries the user identification of the user to be identified; and the business is inquired based on the user identification. The fraud identification database obtains the query result; according to the query result, determines the business fraud identification result of the user to be identified; and sends the business fraud identification result to the terminal.

In one embodiment, the result sending module is also used for confirming that the user to be identified is a business fraud user if it is queried that the business fraud identification database is stored with a business fraud behavior tag corresponding to the user identification.

In one embodiment, the result sending module is also used to obtain the early warning information corresponding to the business fraud behavior label; and send the business fraud identification result and the early warning information to the terminal.

In each of the foregoing embodiments, the device for constructing a business fraud identification database determines the characteristic label of the business fraud event information through the collected key information in the business fraud event information, and then determines the business fraud behavior label of the business fraud event information; according to the business fraud event information To construct a business fraud identification database based on user identifications and business fraud identification tags; in this way, the business fraud identification database constructed can store business fraud identification tags corresponding to multiple user identifications, which is convenient for querying business fraud identification based on the user identification of the user to be identified Database, you can determine whether the business fraud identification database stores business fraud labels corresponding to the user ID, and then determine whether the user to be identified has business fraud; there is no need to analyze and identify a large amount of manually collected data, which simplifies the identification of business fraud The process further improves the identification efficiency of business fraud.

Regarding the specific limitation of the device for constructing the business fraud identification database, please refer to the above limitation on the method of constructing the business fraud identification database, which will not be repeated here. Each module in the device for constructing a business fraud identification database can be implemented in whole or in part by software, hardware, and a combination thereof. The above-mentioned modules may be embedded in the form of hardware or independent of the processor in the computer equipment, or may be stored in the memory of the computer equipment in the form of software, so that the processor can call and execute the operations corresponding to the above-mentioned modules.

In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. 5. The computer equipment includes a processor, a memory, and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide calculation and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database of the computer equipment is used to store data such as user identification and business fraud activity tags of business fraud event information. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for constructing a business fraud identification database is realized.

Those skilled in the art can understand that the structure shown in FIG. 5 is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may Including more or fewer parts than shown in the figure, or combining some parts, or having a different arrangement of parts.

In one embodiment, a computer device is provided, including a memory and a processor, a computer program is stored in the memory, and the processor implements the following steps when executing the computer program: collecting business fraud event information; business fraud event information carries User ID; obtain the key information identifier of the business fraud event information, and extract the corresponding key information from the business fraud event information according to the key information identifier; perform feature extraction on the key information to obtain the feature information of the business fraud event information; obtain The feature code of the feature information, according to the feature code of the feature information, determine the feature tag of the business fraud event information; according to the feature tag of the business fraud event information, query the correspondence between the preset feature tag and the business fraud behavior tag to determine the business fraud event The business fraud label of the information; according to the user identification of the business fraud event information and the business fraud label, the business fraud identification database is constructed.

In one embodiment, the processor further implements the following steps when executing the computer program: collecting multiple business event information; identifying the information content of the business event information, and determining the information type of the business event information according to the information content; The business event information whose information type matches the preset information type is filtered out and used as business fraud event information, and the business fraud event information is uploaded to the blockchain.

In one embodiment, the processor further implements the following steps when executing the computer program: input the feature code of the feature information into the pre-trained feature classification model to obtain the classification probability of the feature information under each preset feature label; The preset feature tag is used as the feature tag of the business fraud event information.

In one embodiment, the processor further implements the following steps when executing the computer program: storing the business fraud behavior label of the business fraud event information in a preset database according to the user identification of the business fraud event information; and identifying the preset database as a business Fraud identification database.

In one embodiment, the processor further implements the following steps when executing the computer program: receiving a business fraud identification request sent by the terminal; carrying the user identification of the user to be identified in the business fraud identification request; querying the business fraud identification database based on the user identification, Obtain the query result; determine the business fraud identification result of the user to be identified according to the query result; send the business fraud identification result to the terminal.

In one embodiment, the processor further implements the following steps when executing the computer program: if the service fraud identification database is queried and stored with a service fraud behavior tag corresponding to the user identification, then the user to be identified is confirmed as a service fraud user.

In one embodiment, the processor further implements the following steps when executing the computer program: acquiring the early warning information corresponding to the business fraud behavior label; and sending the business fraud identification result and the early warning information to the terminal.

In each of the above-mentioned embodiments, the computer device realizes the purpose of storing business fraud identification tags corresponding to multiple user identifications through the constructed business fraud identification database through the computer program running on the processor, so that it is convenient to query the business based on the user identification of the user to be identified. The fraud identification database can determine whether the business fraud identification database stores the business fraud label corresponding to the user ID, and then determine whether the user to be identified has business fraud; there is no need to analyze and identify a large amount of manually collected data, which simplifies business fraud The behavior identification process further improves the identification efficiency of business fraud.

In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented: collecting business fraud event information; the business fraud event information carries a user identification; Obtain the information identifier of the business fraud event information, and extract the corresponding key information from the business fraud event information according to the key information identifier; perform feature extraction on the key information to obtain the characteristic information of the business fraud event information; obtain the characteristics of the characteristic information Encoding, according to the feature code of the feature information, determine the feature label of the business fraud event information; according to the feature label of the business fraud event information, query the correspondence between the preset feature label and the business fraud behavior label to determine the business fraud of the business fraud event information Behavior label: Build a business fraud identification database based on the user identification and business fraud behavior label of the business fraud event information.

In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: collecting multiple business event information; identifying the information content of the business event information, and determining the information type of the business event information according to the information content; The business event information whose information type matches the preset information type is screened out, and used as business fraud event information, and the business fraud event information is uploaded to the blockchain.

In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: input the feature code of the feature information into the pre-trained feature classification model to obtain the classification probability of the feature information under each preset feature label; maximize the classification probability The preset feature tag of is used as the feature tag of business fraud event information.

In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: store the business fraud behavior label of the business fraud event information in a preset database according to the user identification of the business fraud event information; and identify the preset database as Business fraud identification database.

In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: receiving a business fraud identification request sent by the terminal; carrying the user identification of the user to be identified in the business fraud identification request; querying the business fraud identification database based on the user identification , Obtain the query result; determine the business fraud identification result of the user to be identified according to the query result; send the business fraud identification result to the terminal.

In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: if the business fraud identification database is queried and stored in the business fraud identification database, the business fraud behavior tag corresponding to the user identification is confirmed, then the user to be identified is confirmed as a business fraud user.

In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: obtaining the early warning information corresponding to the business fraud behavior label; and sending the business fraud identification result and the early warning information to the terminal.

Optionally, the storage medium involved in this application, such as a computer-readable storage medium, may be non-volatile or volatile.

In each of the foregoing embodiments, the computer-readable storage medium realizes the purpose of storing business fraud identification tags corresponding to multiple user identifications through the constructed business fraud identification database through the computer program stored therein, so as to facilitate the query based on the user identification of the user to be identified The business fraud identification database can determine whether the business fraud identification database stores the business fraud label corresponding to the user ID, and then determine whether the user to be identified has business fraud; there is no need to analyze and identify the large amount of manually collected data, which simplifies the business The fraud identification process further improves the efficiency of business fraud identification.

The blockchain referred to in this application is a new application mode of computer technology such as distributed data storage, point-to-point transmission, consensus mechanism, and encryption algorithm. Blockchain, essentially a decentralized database, is a series of data blocks associated with cryptographic methods. Each data block contains a batch of network transaction information for verification. The validity of the information (anti-counterfeiting) and the generation of the next block. The blockchain can include the underlying platform of the blockchain, the platform product service layer, and the application service layer.

A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium. When the computer program is executed, it may include the processes of the above-mentioned method embodiments. Wherein, any reference to memory, storage, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (Read-Only Memory, ROM), magnetic tape, floppy disk, flash memory, or optical memory. Volatile memory may include random access memory (Random Access Memory, RAM) or external cache memory. As an illustration and not a limitation, RAM can be in many forms, such as Static Random Access Memory (Static Random Access Memory). Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM) etc.

The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be It is considered as the range described in this specification.

The above examples only express several implementation manners of the present application, and the description is relatively specific and detailed, but it should not be understood as a limitation on the scope of the invention patent. It should be pointed out that for those of ordinary skill in the art, without departing from the concept of this application, several modifications and improvements can be made, and these all fall within the protection scope of this application. Therefore, the scope of protection of the patent of this application shall be subject to the appended claims.

Claims

A method for constructing a business fraud identification database, the method comprising:

Collect business fraud event information; the business fraud event information carries a user ID;

Acquiring the key information identifier of the business fraud event information, and extracting corresponding key information from the business fraud event information according to the key information identifier;

Perform feature extraction on the key information to obtain feature information of the business fraud event information;

Acquiring the feature code of the feature information, and determining the feature tag of the business fraud event information according to the feature code of the feature information;

According to the feature tag of the business fraud event information, query the correspondence between the preset feature tag and the business fraud tag to determine the business fraud tag of the business fraud event information;

According to the user identification and the business fraud behavior label of the business fraud event information, a business fraud identification database is constructed.
The method according to claim 1, wherein said collecting business fraud event information comprises:

Collect multiple business event information;

Identifying the information content of the business event information, and determining the information type of the business event information according to the information content;

Filter out the business event information whose information type matches the preset information type from multiple business event information, and use it as business fraud event information;

Upload the business fraud event information to the blockchain.
The method according to claim 1, wherein said determining the feature tag of the business fraud event information according to the feature code of the feature information comprises:

Input the feature code of the feature information into a pre-trained feature classification model to obtain the classification probability of the feature information under each preset feature label;

The preset feature label with the largest classification probability is used as the feature label of the business fraud event information.
The method according to claim 1, wherein the constructing a business fraud identification database according to the user identification of the business fraud event information and the business fraud behavior label comprises:

Storing the business fraud behavior label of the business fraud event information in a preset database according to the user identification of the business fraud event information;

The preset database is identified as the business fraud identification database.
The method according to any one of claims 1 to 4, wherein the method further comprises:

Receiving a business fraud identification request sent by the terminal; the business fraud identification request carries the user identification of the user to be identified;

Query the business fraud identification database according to the user ID to obtain the query result;

Determine, according to the query result, a business fraud identification result of the user to be identified;

Send the recognition result of the business fraud to the terminal.
The method according to claim 5, wherein the determining the result of business fraud identification of the user to be identified according to the query result comprises:

If it is found that the business fraud identification database stores a business fraud behavior tag corresponding to the user identification, then it is confirmed that the user to be identified is a business fraud user.
The method according to claim 6, wherein after confirming that the user to be identified is a business fraud user, the method further comprises:

Obtain the early warning information corresponding to the business fraud behavior label;

The sending the identification result of the business fraud behavior to the terminal includes:

Sending the business fraud identification result and the early warning information to the terminal.
A device for constructing a business fraud identification database, wherein the device includes:

An information collection module, used to collect business fraud event information; said business fraud event information carries user identification;

An information extraction module, configured to obtain the key information identifier of the business fraud event information, and extract corresponding key information from the business fraud event information according to the key information identifier;

The information determining module is used for feature extraction of the key information to obtain the feature information of the business fraud event information;

A characteristic label determination module, configured to obtain the characteristic code of the characteristic information, and determine the characteristic label of the business fraud event information according to the characteristic code of the characteristic information;

The behavior label determination module is configured to query the correspondence between the preset characteristic label and the business fraud behavior label according to the characteristic label of the business fraud event information, and determine the business fraud behavior label of the business fraud event information;

The database construction module is used to construct a business fraud identification database according to the user identification of the business fraud event information and the business fraud behavior label.
A computer device includes a memory and a processor, the memory stores a computer program, wherein the processor implements the following method when the computer program is executed:

Collect business fraud event information; the business fraud event information carries a user ID;

Acquiring the key information identifier of the business fraud event information, and extracting corresponding key information from the business fraud event information according to the key information identifier;

Perform feature extraction on the key information to obtain feature information of the business fraud event information;

Acquiring the feature code of the feature information, and determining the feature tag of the business fraud event information according to the feature code of the feature information;

According to the feature tag of the business fraud event information, query the correspondence between the preset feature tag and the business fraud tag to determine the business fraud tag of the business fraud event information;

According to the user identification and the business fraud behavior label of the business fraud event information, a business fraud identification database is constructed.
The computer device according to claim 9, wherein, when performing the collection of business fraud event information, it comprises:

Collect multiple business event information; identify the information content of the business event information, and determine the information type of the business event information according to the information content;

Filter out the business event information whose information type matches the preset information type from multiple business event information, and use it as business fraud event information;

Upload the business fraud event information to the blockchain.
9. The computer device according to claim 9, wherein the execution of the feature encoding based on the feature information to determine the feature tag of the business fraud event information comprises:

Input the feature code of the feature information into a pre-trained feature classification model to obtain the classification probability of the feature information under each preset feature label;

The preset feature label with the largest classification probability is used as the feature label of the business fraud event information.
The computer device according to any one of claims 9 to 11, wherein the processor is further configured to implement:

Receiving a business fraud identification request sent by the terminal; the business fraud identification request carries the user identification of the user to be identified;

Query the business fraud identification database according to the user ID to obtain the query result;

Determine, according to the query result, a business fraud identification result of the user to be identified;

Send the recognition result of the business fraud to the terminal.
11. The computer device according to claim 12, wherein the execution of determining the result of business fraud identification of the user to be identified according to the query result comprises:

If it is queried that the business fraud identification database stores a business fraud behavior tag corresponding to the user identification, then it is confirmed that the user to be identified is a business fraud user.
The computer device according to claim 13, wherein, after confirming that the user to be identified is a business fraud user, the processor is further configured to: obtain early warning information corresponding to the business fraud behavior label;

When performing the sending of the recognition result of the business fraud behavior to the terminal, it includes:

Sending the business fraud identification result and the early warning information to the terminal.
A computer-readable storage medium having a computer program stored thereon, wherein the computer program is executed by a processor to implement the following method:

Collect business fraud event information; the business fraud event information carries a user ID;

Acquiring the key information identifier of the business fraud event information, and extracting corresponding key information from the business fraud event information according to the key information identifier;

Perform feature extraction on the key information to obtain feature information of the business fraud event information;

Acquiring the feature code of the feature information, and determining the feature tag of the business fraud event information according to the feature code of the feature information;

According to the feature tag of the business fraud event information, query the correspondence between the preset feature tag and the business fraud tag to determine the business fraud tag of the business fraud event information;

According to the user identification and the business fraud behavior label of the business fraud event information, a business fraud identification database is constructed.
The computer-readable storage medium according to claim 15, wherein, when performing the collection of business fraud event information, it comprises:

Collect multiple business event information; identify the information content of the business event information, and determine the information type of the business event information according to the information content;

Filter out the business event information whose information type matches the preset information type from multiple business event information, and use it as business fraud event information;

Upload the business fraud event information to the blockchain.
15. The computer-readable storage medium according to claim 15, wherein the execution of the feature encoding according to the feature information to determine the feature tag of the business fraud event information comprises:

Input the feature code of the feature information into a pre-trained feature classification model to obtain the classification probability of the feature information under each preset feature label;

The preset feature label with the largest classification probability is used as the feature label of the business fraud event information.
The computer-readable storage medium according to any one of claims 15 to 17, wherein, when the computer program is executed by a processor, it is also used to implement:

Receiving a business fraud identification request sent by the terminal; the business fraud identification request carries the user identification of the user to be identified;

Query the business fraud identification database according to the user ID to obtain the query result;

Determine, according to the query result, a business fraud identification result of the user to be identified;

Send the recognition result of the business fraud to the terminal.
18. The computer-readable storage medium according to claim 18, wherein the execution of determining the business fraud identification result of the user to be identified according to the query result comprises:

If it is found that the business fraud identification database stores a business fraud behavior tag corresponding to the user identification, then it is confirmed that the user to be identified is a business fraud user.
The computer-readable storage medium according to claim 19, wherein, after confirming that the user to be identified is a business fraud user, when the computer program is executed by the processor, the computer program is further used to implement: obtaining a label related to the business fraud Corresponding warning information;

When performing the sending of the recognition result of the business fraud behavior to the terminal, it includes:

Sending the business fraud identification result and the early warning information to the terminal.