Data processing method and device, computer readable medium and electronic equipment

Abstract

The application belongs to the technical field of computers, and particularly relates to a data processing method, a data processing device, a computer readable medium and an electronic device. The method comprises the following steps: obtaining service data to be processed, wherein the service data comprises service text and service attributes associated with the service text; obtaining keyword data used for abnormality detection of the service data, wherein the keyword data comprises keyword text and keyword attributes associated with the keyword text; performing matching detection on the keyword text and the service text to determine whether the keyword text is contained in the service text; if the keyword text and the service text are successfully matched, combining the service data and the keyword data to form matching data; and extracting target data used for evaluating abnormality risk of the service data from the matching data according to the service attributes and the keyword attributes. The application can improve the data processing efficiency and data recognition accuracy of risk detection.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of computer technology, and particularly relates to a data processing method, a data processing device, a computer readable medium and an electronic device. BACKGROUND

[0002] Various types of financial institutions such as banks and insurance companies need to strictly supervise and review the financial business involved in the operation process. For example, when a large transaction behavior or a suspicious transaction behavior is found, the transaction content should be risk assessed, and if necessary, reported to the superior supervisory agency. However, the existing due diligence method mostly depends on manual review, and there is a problem of low processing efficiency and easy omission of risks. SUMMARY

[0003] The present application aims to provide a data processing method, a data processing device, a computer readable medium and an electronic device, which at least partly overcome the technical problems of low processing efficiency and poor accuracy in the related art.

[0004] Other characteristics and advantages of the present application will become apparent from the following detailed description, or will be learned by practice of the present application.

[0005] According to an aspect of an embodiment of the present application, a data processing method is provided, which includes: acquiring business data to be processed, the business data including business text and business attributes associated with the business text, the business attributes being used to represent data collection sources and data collection times of the business data; acquiring keyword data used for abnormality detection of the business data, the keyword data including keyword text and keyword attributes associated with the keyword text, the keyword attributes being used to represent keyword types and abnormality risk levels of the keyword data; performing matching detection on the keyword text and the business text to determine whether the keyword text is contained in the business text; if the keyword text and the business text are successfully matched, combining the business data and the keyword data to form matching data; and extracting target data used to evaluate abnormality risks of the business data from the matching data according to the business attributes and the keyword attributes.

[0006] According to an aspect of some embodiments of the present application, a data processing apparatus is provided, which comprises: a service data obtaining module configured to obtain service data to be processed, the service data comprising service text and service attributes associated with the service text, the service attributes being used to represent data collection sources and data collection times of the service data; a keyword data obtaining module configured to obtain keyword data used for abnormality detection of the service data, the keyword data comprising keyword text and keyword attributes associated with the keyword text, the keyword attributes being used to represent keyword types and abnormality risk levels of the keyword data; a matching detection module configured to perform matching detection on the keyword text and the service text to determine whether the keyword text is contained in the service text; a data combination module configured to combine the service data and the keyword data to form matching data if the keyword text and the service text are successfully matched; and a data extraction module configured to extract target data used for evaluating abnormality risks of the service data from the matching data.

[0007] In some embodiments of the present application, based on the above technical solutions, the data extraction module comprises: a quantization processing module configured to perform quantization processing on the matching data according to the service attributes and the keyword attributes to obtain quantization data corresponding to the matching data, the quantization data being numerical data comprising one or more numerical fields; a data sorting module configured to sort the matching data according to values of the numerical fields in the quantization data to obtain a data list arranged in order; and a data selection module configured to select target data used for evaluating abnormality risks of the service data from the data list according to the arrangement order.

[0008] In some embodiments of the present application, based on the above technical solutions, the quantization processing module comprises: a field extraction module configured to extract a first service field related to the service attributes, a second service field related to the keyword attributes, and a third service field related to statistical information of the matching data from the matching data; a field quantization module configured to perform quantization processing on the first service field, the second service field and the third service field respectively according to field types of the service fields to obtain numerical fields corresponding to the service fields; and a field combination module configured to combine the numerical fields to obtain the quantization data corresponding to the matching data.

[0009] In some embodiments of the present application, based on the above technical solution, the first service field comprises a time field for representing the collection time of the service data, the second service field comprises a level field for representing the abnormal risk level of the keyword data, and the third service field comprises a frequency field for representing the occurrence frequency of the keyword text in the matching data.

[0010] In some embodiments of the present application, based on the above technical solution, the data sorting module comprises: a weight obtaining module configured to obtain quantization weights corresponding to each numerical field in the quantization data; a field weighting module configured to perform weighted summation on the numerical fields according to the quantization weights to obtain a selection probability of the matching data; and a probability sorting module configured to sort the matching data according to the selection probability.

[0011] In some embodiments of the present application, based on the above technical solution, the weight obtaining module comprises: a sample obtaining module configured to obtain quantization data samples obtained by quantizing historical service data and data labels obtained by evaluating the abnormal risk of the historical service data; a sample combination module configured to combine the quantization data samples and the data labels into training samples; and a weight prediction module configured to predict quantization weights corresponding to each numerical field in the quantization data according to the training samples.

[0012] In some embodiments of the present application, based on the above technical solution, the weight prediction module comprises: a model obtaining module configured to obtain a weight prediction model for predicting the weights of the quantization data; an iterative training module configured to iteratively train the weight prediction model according to the training samples; and a model prediction module configured to predict the quantization weights corresponding to each numerical field in the quantization data through the trained weight prediction model.

[0013] In some embodiments of the present application, based on the above technical solution, the weight prediction module comprises: a matrix combination module configured to combine the quantization data samples and the data labels in the training samples into a data sample matrix and a data label matrix according to the data types; and a matrix operation module configured to perform matrix operation on the data sample matrix and the data label matrix to obtain a weight matrix, wherein the weight matrix comprises quantization weights corresponding to each numerical field in the quantization data.

[0014] In some embodiments of the present application, based on the above technical solution, the matrix combination module is configured to: perform screening processing on the training samples to obtain a set number of relevant samples having data correlation with the business data; and respectively group the quantized data samples and the data labels in the relevant samples according to data types to form a data sample matrix and a data label matrix.

[0015] In some embodiments of the present application, based on the above technical solution, the data sorting module is configured to: obtain a sorting priority corresponding to each numerical field in the quantized data; and sort the matching data according to the sorting priority and the value of each numerical field.

[0016] In some embodiments of the present application, based on the above technical solution, the data processing apparatus further comprises: a hyperlink adding module configured to add a hyperlink text related to the data source of the business data in the data list; and a cloud map display module configured to display a keyword cloud map related to the data source of the business data in response to a triggering operation on the hyperlink text.

[0017] In some embodiments of the present application, based on the above technical solution, the data processing apparatus further comprises: a text selecting module configured to extract a keyword text corresponding to a selected keyword type from the target data in response to a selection operation on the keyword type; and a message generating module configured to fill the keyword text into a preset message template to generate an evaluation message for evaluating the abnormal risk of the business data.

[0018] According to an aspect of an embodiment of the present application, there is provided a computer readable medium having stored thereon a computer program, which, when executed by a processor, implements the data processing method in the above technical solution.

[0019] According to an aspect of an embodiment of the present application, there is provided an electronic device, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute the data processing method in the above technical solution by executing the executable instructions.

[0020] According to an aspect of an embodiment of the present application, there is provided a computer program product or computer program, comprising computer instructions stored in a computer readable storage medium. A processor of a computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions, so that the computer device executes the data processing method in the above technical solution.

[0021] In the technical solution provided in the embodiments of the present application, the business text in the business data is matched and detected by the keyword text, so that the business text with abnormal risk can be quickly identified, and after the business text with abnormal risk and the corresponding keyword data are matched to form matching data, the target data used for evaluating the abnormal risk is extracted from the matching data by the business attribute and the keyword attribute. The embodiments of the present application can improve the processing efficiency of risk detection on the business data, and can improve the accuracy of data identification.

[0022] It should be understood that the foregoing general description and the following detailed description are only exemplary and explanatory, and are not limiting to the present application. BRIEF DESCRIPTION OF DRAWINGS

[0023] The drawings incorporated in the specification and constituting a part of the specification illustrate embodiments consistent with the present application and, together with the specification, serve to explain the principles of the present application. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0024] Figure 1 An exemplary system architecture block diagram to which the technical solution of the present application is applied is schematically shown.

[0025] Figure 2 An architecture block diagram of a data processing system in an embodiment of the present application is schematically shown.

[0026] Figure 3 A step flowchart of a data processing method in an embodiment of the present application is schematically shown.

[0027] Figure 4 A table diagram schematically showing storage of business data in an application scenario by an embodiment of the present application is schematically shown.

[0028] Figure 5 A process diagram schematically showing matching detection of keyword text to obtain matching data in an application scenario by an embodiment of the present application is schematically shown.

[0029] Figure 6 A method step flowchart of extracting target data in an embodiment of the present application is schematically shown.

[0030] Figure 7 A data table diagram schematically showing training samples in an embodiment of the present application is schematically shown.

[0031] Figure 8 An interface diagram schematically showing a data list in an embodiment of the present application is schematically shown.

[0032] Figure 9A structural block diagram of a data processing apparatus provided by an embodiment of the present application is shown schematically.

[0033] Figure 10 A structural block diagram of a computer system of an electronic device suitable for implementing an embodiment of the present application is shown schematically. DETAILED DESCRIPTION

[0034] Example implementations are now described with reference to the drawings. Example implementations can, however, be implemented in many different forms and should not be construed as limited to the examples set forth herein; rather, these implementations are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the example implementations to those skilled in the art.

[0035] Moreover, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of embodiments of the application. One skilled in the relevant art will recognize, however, that the

[0036] The block diagrams in the drawings show only the functional entities and do not necessarily correspond to physically separate entities. That is, the functional entities can be implemented in software, or in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0037] The flow diagrams shown in the drawings are merely examples and do not necessarily include all of the content and operations / steps, nor do they necessarily have to be performed in the order described. For example, some operations / steps can be broken down further, and some operations / steps can be combined or partially combined, so the actual order of performance can be changed according to actual conditions.

[0038] Figure 1 An example system architecture block diagram to which the technical solution of the present application is applied is shown schematically.

[0039] As Figure 1As shown, the system architecture 100 can include a terminal device 110, a network 120 and a server 130. The terminal device 110 can include various electronic devices such as a smartphone, a tablet computer, a notebook computer, a desktop computer, etc. The server 130 can be a standalone physical server, a server cluster composed of multiple physical servers or a distributed system, or a cloud server providing cloud computing services. The network 120 can be a communication medium of various connection types capable of providing a communication link between the terminal device 110 and the server 130, for example, a wired communication link or a wireless communication link.

[0040] According to implementation needs, the system architecture in the embodiments of the present application can have any number of terminal devices, networks and servers. For example, the server 130 can be a server group composed of multiple server devices. In addition, the technical solutions provided by the embodiments of the present application can be applied to the terminal device 110, or can be applied to the server 130, or can be jointly implemented by the terminal device 110 and the server 130, and the present application does not make special limitations on this.

[0041] For example, a user initiates a payment, transfer or other business request to a bank or other financial institution through the terminal device 110, and the server 130 running at the financial institution can handle the corresponding business according to the specified business rules after receiving the business request initiated by the user through the network 120. At the same time, the server 130 can supervise and predict the risks of the business data generated in the business handling process, so as to discover abnormal transaction behaviors in time.

[0042] Figure 2 The architecture block diagram of a data processing system in an embodiment of the present application is shown, which can be implemented through an application program for risk prediction installed on the terminal device 110 or the server 130.

[0043] As Figure 2As shown, the data processing system 200 can mainly include a keyword matching module 210, a keyword analysis module 220 and a keyword application module 230. Among them, the keyword matching module first extracts the text information from the business information table 240, and can obtain the keyword data from the keyword database 250, and after matching detection of the text information by using the keyword data, a data list composed of matching results is obtained, and then the data list is transmitted to the keyword analysis model 220. The keyword analysis module 220 can use linear mode or nonlinear mode to clean the data list, and extract the main risk text information with abnormal risk. The keyword application module 230 can further process the main risk text information extracted by the keyword analysis module 220 according to the actual application requirement, and obtain the data processing result in the specific application scene. For example, a data report or early warning information for risk prompt of the business data with abnormal risk can be finally formed, and the like.

[0044] In some embodiments of the present application, the data processing system 200 can be implemented by an artificial intelligence-based machine learning model.

[0045] Artificial intelligence (AI) is to use digital computers or digital computer controlled machines to simulate, extend and expand human intelligence, perceive environment, acquire knowledge and use knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technology of computer science, which tries to understand the essence of intelligence and produce a new intelligent machine that can react in a similar way to human intelligence. Artificial intelligence is to study the design principles and implementation methods of various intelligent machines, so that the machine has the functions of perception, reasoning and decision-making.

[0046] Artificial intelligence technology is a comprehensive discipline, involving a wide range of fields, both hardware and software technologies. Artificial intelligence basic technologies generally include sensors, special artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation / interaction system, mechatronics and other technologies. Artificial intelligence software technology mainly includes computer vision technology, speech processing technology, natural language processing technology and machine learning / deep learning and other fields.

[0047] Natural language processing (NLP) is an important direction in the field of computer science and artificial intelligence. It studies various theories and methods that can realize effective communication between people and computers using natural language. Natural language processing is a science that integrates linguistics, computer science and mathematics. Therefore, the research in this field will involve natural language, i.e. the language used in daily life, so it is closely related to the study of linguistics. Natural language processing technology usually includes text processing, semantic understanding, machine translation, robot question and answer, knowledge graph and other technologies.

[0048] Machine learning (ML) is a multi-disciplinary subject that involves probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory and other disciplines. It is a specialized study of how computers simulate or implement human learning behavior to acquire new knowledge or skills, reorganize existing knowledge structure and continuously improve their performance. Machine learning is the core of artificial intelligence and the fundamental approach to making computers intelligent. Its applications are widespread in various fields of artificial intelligence. Machine learning and deep learning usually include artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and rule-based learning.

[0049] In some embodiments of the present application, the keyword database and the business information table can be reliably stored in a decentralized manner through blockchain technology.

[0050] Blockchain is a new application mode of distributed data storage, peer-to-peer transmission, consensus mechanism, encryption algorithm and other computer technologies. Blockchain, in essence, is a decentralized database, a series of data blocks associated using cryptographic methods, each containing a batch of network transaction information for verifying the validity of the information (anti-fake) and generating the next block. Blockchain can include blockchain underlying platform, platform product service layer and application service layer.

[0051] The blockchain underlying platform can include user management, basic services, smart contracts, and operation detection processing modules. Among them, the user management module is responsible for the identity information management of all blockchain participants, including maintaining public and private key generation (account management), key management, and user real identity and blockchain address correspondence maintenance (permission management), etc., and under authorization, supervises and audits the transaction of certain real identities, provides risk control rule configuration (risk audit); the basic service module is deployed on all blockchain node devices to verify the validity of business requests, and record to the storage after consensus for valid requests, for a new business request, the basic service first interface adaptation analysis and authentication processing (interface adaptation), then encrypt the business information through the consensus algorithm (consensus management), after encryption, the complete and consistent transmission to the shared ledger (network communication), and record storage; the smart contract module is responsible for contract registration and issuance, contract triggering and contract execution, developers can define contract logic through a certain programming language, publish to the blockchain (contract registration), according to the logic of the contract terms, call the key or other event triggers to execute, complete the contract logic, and also provide contract upgrade and cancellation functions; the operation detection module is mainly responsible for the deployment, configuration modification, contract setting, cloud adaptation in the product release process, and the real-time state visualization output in the product running, such as: alarm, detection network situation, detection node device health state, etc.

[0052] The platform product service layer provides basic capabilities and implementation frameworks for typical applications. Developers can add business features based on these basic capabilities to implement business logic on the blockchain.

[0053] The data processing method, data processing apparatus, computer readable medium, and electronic device provided by the present application will be described in detail below in conjunction with specific embodiments.

[0054] Figure 3 A step flowchart of a data processing method in an embodiment of the present application is shown, which can be executed by a terminal device or a server, or jointly executed by a terminal device and a server. The present embodiment takes the data processing method executed by a terminal device as an example for description. As shown in Figure 3 The data processing method can mainly include the following steps S310-S350.

[0055] Step S310: Obtain the to-be-processed business data, the business data including a business text and a business attribute associated with the business text, the business attribute being used to indicate a data collection source and a data collection time of the business data.

[0056] Step S320: Obtain keyword data used for anomaly detection of the service data, the keyword data including keyword text and keyword attributes associated with the keyword text, the keyword attributes being used to represent keyword types and anomaly risk levels of the keyword data.

[0057] Step S330: Perform matching detection on the keyword text and the service text to determine whether the keyword text is contained in the service text.

[0058] Step S340: If the keyword text and the service text are successfully matched, combine the service data and the keyword data to form matching data.

[0059] Step S350: Extract target data used for evaluating the anomaly risk of the service data from the matching data according to the service attributes and the keyword attributes.

[0060] In the data processing method provided in the embodiments of the present application, the service text in the service data is matched and detected through the keyword text, so that the service text with an abnormal risk can be quickly identified, and after the service text is combined with the corresponding keyword data to form matching data, target data used for evaluating the abnormal risk is extracted from the matching data according to the service attributes and the keyword attributes. The method can improve the processing efficiency of risk detection of the service data, and can improve the accuracy of data identification.

[0061] The various method steps in the data processing method of the embodiments of the present application will be described in detail below in combination with specific application scenarios.

[0062] In step S310, obtain service data to be processed, the service data including service text and service attributes associated with the service text, the service attributes being used to represent data collection sources and data collection times of the service data.

[0063] In an embodiment of the present application, the service data to be processed can be transaction data obtained after screening and information collection of large-amount transactions or suspicious transactions found in monitoring, and can specifically include service information and user information generated by a user when performing a financial service, and the like. The service text is text content generated in a service process, and can include, for example, a name of a transaction and descriptive words for explaining a transaction content. The service attributes are attribute contents associated with the service text, and can include fields representing multiple attribute dimensions such as data collection sources and data collection times.

[0064] For example, a transfer transaction between two or more users generates business data that may include the account numbers of the parties involved, the transaction time, the transfer amount, and a transfer note. The transfer note or business description is the business text, while the associated account number or account name is a business attribute used to indicate the source of the data, and the associated transaction time is a business attribute used to indicate when the data was collected.

[0065] In one embodiment of the present application, the business data may be a data table formed by combining business text and business attributes according to specified dimensions. Figure 4 A schematic diagram of a table for storing business data in an application scenario in an embodiment of the present application is shown, and the table is a business information table. Figure 4 As shown in the table, the first column is the serial number id, the second column is the matched text content word, the third column is the customer name whos to which the text information belongs, and the fourth column is the time time when the text information appears.

[0066] In step S320 , keyword data for performing anomaly detection on business data is obtained. The keyword data includes keyword text and keyword attributes associated with the keyword text. The keyword attributes are used to represent the keyword type and anomaly risk level of the keyword data.

[0067] In one embodiment of the present application, the keyword data may be a portion of database entries extracted from a keyword database for use in detecting anomalies in business data. The keyword database may be a database formed by collecting and screening historical business data.

[0068] Keyword data includes keyword text and one or more keyword attributes associated with the keyword text. The keyword attributes are used to indicate the keyword type, abnormal risk level and other attribute information related to the abnormal risk of business data for each keyword data. Keyword types may include, for example, white text and abnormal text, wherein white text indicates that the corresponding keyword text is a normal text without abnormal risk, while abnormal text indicates that the keyword text is a text with a specified type of abnormal risk. The abnormal risk level is used to indicate the degree of abnormal risk of keyword texts corresponding to different keyword types. The abnormal risk level may include attribute text that qualitatively measures the degree of abnormal risk, such as high risk level, medium risk level, and low risk level. The abnormal risk level may also include attribute values ​​that quantitatively measure the degree of abnormal risk, such as continuous or segmented values. The larger the value, the higher the degree of abnormal risk.

[0069] In an embodiment of the present application, the keyword data can include a combination of fields composed of a keyword text field, a keyword type field and an abnormal risk level field. Taking the application scenario of due diligence as an example, the keyword type can include white text and abnormal text. For example, the keyword data can include the following combinations of fields:

white text, XXX, a

abnormal text, YYY, b

abnormal text, ZZZ, c

[0070] Step S330: Matching detection of the keyword text and the business text to determine whether the keyword text is contained in the business text.

[0071] The method of matching detection of the keyword text and the business text can include obtaining the text length of the keyword text, taking the text length of the keyword text as a sliding window, sampling text strings on the business text, and sequentially matching the sampled text strings with the keyword text. When at least one text string in the business text matches the keyword text successfully, it can be determined that the keyword text is contained in the business text.

[0072] In an embodiment of the present application, through matching detection, it can not only be determined whether the keyword text is contained in the business text, but also the frequency of occurrence of the keyword text in the business text can be counted. The higher the frequency of occurrence of the keyword text, the greater the influence of the keyword text on the abnormal detection result of the business text.

[0073] Step S340: If the keyword text and the business text match successfully, the business data and the keyword data are combined to form matching data.

[0074] For the keyword text and the business text that match successfully, the business data containing the business text and the keyword data containing the keyword text can be combined to form corresponding matching data. In an embodiment of the present application, the matching data can be complete data obtained by combining all fields in the business data and the keyword data, or the matching data can be incomplete data obtained by combining part of the fields extracted from the business data and the keyword data.

[0075] Figure 5 A process diagram of matching detection of keyword text to obtain matching data in an application scenario is shown.

[0076] As Figure 5As shown, in one matching round, the keyword data 510 to be matched may be a field combination

weather, white text, 0

[0077] The business texts are extracted one by one from the business information table 520 storing the business data, and are input into the keyword matching module 530 together with the keyword text, so as to determine by the keyword matching module 530 whether the business text contains the same text string as the keyword text. For example, the business text extracted in the current matching round is "Today the weather is really good", which contains the keyword text "weather", so it can be determined that the matching is successful. Conversely, if the business text does not contain the same text string as the keyword text, it is determined that the matching fails.

[0078] When the matching is successful, the keyword data 510 can be combined with the business data matched successfully to obtain matching data 540. For example, the business data is a field combination

today the weather is really good, Zhang San, 2020-1-2

weather, white text, 0, today the weather is really good, Zhang San, 2020-1-2

[0079] When the matching fails, the keyword data 510 can continue to be matched with the next business text extracted from the business information table 520. If all the business texts in the business information table 520 have completed the matching detection with the keyword data 510, a new set of keyword data can be selected and the next round of matching detection can be started.

[0080] After the matching detection is completed and the matching data is obtained, the target data can be further extracted therefrom, so as to evaluate the abnormal risk of the business data according to the target data.

[0081] Figure 6 A method step flowchart for extracting the target data in one embodiment of the application is shown. As shown in Figure 6 On the basis of the above embodiment, the step S350 of extracting the target data for evaluating the abnormal risk of the business data from the matching data according to the business attribute and the keyword attribute may include the following steps S610 to S630.

[0082] Step S610: quantifying the matching data according to the business attribute and the keyword attribute, to obtain quantized data corresponding to the matching data, the quantized data being numerical data including one or more numerical fields.

[0083] Step S620: sorting the matching data according to the values of the numerical fields in the quantized data, to obtain a data list in an ordered arrangement.

[0084] Step S630: selecting target data for evaluating the abnormal risk of the business data from the data list according to the arrangement order.

[0085] In the embodiments of the present application, the matching data is sorted by obtaining quantized data corresponding to multiple dimensions of the business attribute and the keyword attribute, and the target data can be selected in a manner of comparing quantized numerical values, so that the data selection efficiency of the target data is improved, and the reliability of the abnormal risk evaluation is improved.

[0086] The steps of extracting the target data will be described in detail in combination with specific embodiments.

[0087] In step S610, the matching data is quantized according to the business attribute and the keyword attribute, to obtain quantized data corresponding to the matching data, the quantized data being numerical data including one or more numerical fields.

[0088] In an embodiment of the present application, the method of quantizing the matching data in step S610 can include: extracting a first business field related to the business attribute, a second business field related to the keyword attribute, and a third business field related to statistical information of the matching data from the matching data; quantizing the first business field, the second business field, and the third business field according to the field types of the respective business fields, to obtain numerical fields corresponding to the respective business fields; and combining the numerical fields to obtain the quantized data corresponding to the matching data.

[0089] According to the preset rules, the business fields related to the business attribute and the keyword attribute in the matching data and the statistical information of the matching data are quantized respectively, to obtain numerical fields corresponding to the multiple business fields, and then the multiple numerical fields are combined to obtain the quantized data corresponding to the matching data.

[0090] In an embodiment of the present application, the first business field includes a time field for indicating the collection time of the business data, the second business field includes a level field for indicating the abnormal risk level of the keyword data, and the third business field includes a frequency field for indicating the occurrence frequency of the keyword text in the matching data.

[0091] For the first business field, a time period division manner can be adopted to quantize to obtain a corresponding numerical field. For example, according to the data collection time in the business attribute, the distance between the collection time of the business data and the current time can be determined, and then the time distance is compared with the preset unit time length to obtain the corresponding quantized numerical value. For example, when the distance from the current time is within 1 month, the corresponding quantized numerical value is determined as 1; when the distance from the current time is between 1 month and 2 months, the corresponding quantized numerical value can be determined as 2; and so on.

[0092] For the second business field, if the abnormal risk level is numerical data in quantitative representation, the value thereof can be directly selected as the quantized numerical field, or the normalized value or the value after other processing can be used as the quantized numerical field. If the abnormal risk level is a text field in qualitative representation, the qualitative level can be mapped to a corresponding quantized numerical value according to a preset mapping relationship. For example, the abnormal risk level is qualitatively divided into three levels of high, medium and low, and then the three levels can be mapped to quantized numerical values of 2, 1 and 0, respectively.

[0093] For the third business field, the value of the occurrence frequency can be directly selected as the quantized numerical field, or the normalized value or the value after other processing can be used as the quantized numerical field.

[0094] In step S620, the matching data is sorted according to the values of the numerical fields in the quantized data to obtain an ordered data list.

[0095] In an embodiment of the present application, the method of sorting the matching data can include: obtaining quantized weights corresponding to the numerical fields in the quantized data; performing weighted summation on the numerical fields according to the quantized weights to obtain a selection probability of the matching data; and sorting the matching data according to the selection probability.

[0096] For example, the quantized data includes a first quantized field L corresponding to the abnormal risk level, a second quantized field C corresponding to the keyword occurrence frequency, and a third quantized field T corresponding to the data collection time. The weighted summation of the respective quantized fields according to the respective quantized weights β 1, β 2 and β 3 can obtain a selection probability P as:

[0097]

[0098] In an embodiment of the present application, the quantized weight can be a fixed numerical value preset according to the processing experience of historical business data, or can be a dynamic numerical value dynamically determined in the abnormal risk evaluation process.

[0099] In one embodiment of the present application, a method for obtaining quantitative weights corresponding to each numerical field in the quantitative data may include: obtaining quantitative data samples obtained by quantitatively processing historical business data, and obtaining data labels obtained by performing abnormal risk assessment on the historical business data; combining the quantitative data samples and data labels into training samples; and predicting the quantitative weights corresponding to each numerical field in the quantitative data based on the training samples.

[0100] Figure 7 Schematic diagram of a data table of training samples in one embodiment of the present application is shown. Figure 7 As shown, the data table of the training sample includes two parts: data label 701 and quantitative data sample 702. Data label 701 indicates whether the historical business data, determined after abnormal risk assessment, has abnormal risk. For example, when the data label is 1, it means that the data sample has abnormal risk, and when the data label is 0, it means that the data sample does not have abnormal risk. Quantified data sample 702 is composed of quantitative fields of multiple dimensions, such as a first quantitative field L corresponding to the abnormal risk level, a second quantitative field C corresponding to the frequency of keyword occurrence, and a third quantitative field T corresponding to the data collection time.

[0101] In one embodiment of the present application, different quantization weight calculation methods can be selected according to the data volume of the training sample. For example, when the data volume of the training sample is greater than the set data volume threshold, the quantization weight can be determined by using the training weight prediction model method; and when the data volume of the training sample is less than or equal to the set data volume threshold, the quantization weight can be determined by using the matrix operation method.

[0102] In one embodiment of the present application, a method for predicting quantization weights corresponding to various numerical fields in quantization data based on training samples may include: obtaining a weight prediction model for weight prediction of quantization data; iteratively training the weight prediction model based on the training samples; and predicting the quantization weights corresponding to various numerical fields in the quantization data using the trained weight prediction model. In an embodiment of the present application, the weight prediction model may be a linear model or a nonlinear model for mapping input data to predict quantization weights. The weight prediction model may be trained using a gradient descent method to improve its weight prediction effect.

[0103] In one embodiment of the present application, a method for predicting quantization weights corresponding to each numerical field in quantized data based on training samples may include: respectively organizing the quantized data samples and data labels in the training samples into a data sample matrix and a data label matrix according to data types; performing matrix operations on the data sample matrix and the data label matrix to obtain a weight matrix, the weight matrix including the quantization weights corresponding to each numerical field in the quantized data.

[0104] by Figure 7 Taking the training sample shown as an example, the data labels 701 can form a data label matrix Y. Y = [1 ... 0].

[0105] The quantized data samples 702 may form a data sample matrix X.

[0106]

[0107] The data label matrix Y and the data sample matrix X can be operated using the following matrix operation formula to obtain the weight matrix A.

[0108]

[0109] In one embodiment of the present application, a method for forming a data sample matrix and a data label matrix based on data type for quantized data samples and data labels in training samples may include: screening the training samples to obtain a set number of related samples that have data relevance to business data; and forming a data sample matrix and a data label matrix based on data type for quantized data samples and data labels in the related samples. When the amount of training sample data is large, the matrix operation process will consume a large amount of computing resources. By screening the training samples, the embodiment of the present application can reduce the amount of matrix operation and improve data processing efficiency.

[0110] In one embodiment of the present application, the method of sorting the matching data according to the values ​​of each numerical field in the quantitative data in step S620 may include: obtaining a sorting priority corresponding to each numerical field in the quantitative data; and sorting the matching data according to the sorting priority and the values ​​of each numerical field. For example, in the embodiment of the present application, the matching data is first sorted according to the first numerical field with the highest priority among the sorting priorities. For matching data with the same first numerical field, the matching data is then sorted according to the second numerical field with the second highest sorting priority, and so on, until the sorting comparison of each numerical field under all sorting priorities is completed.

[0111] In one embodiment of the present application, after sorting the matching data based on the values ​​of each numerical field in the quantitative data to obtain an ordered data list, hyperlink text related to the data source of the business data can be added to the data list; in response to a triggering operation on the hyperlink text, a keyword cloud chart related to the data source of the business data is displayed. Based on the keyword cloud chart, basic information about the data source (such as a specific transaction user) can be intuitively viewed.

[0112] In step S630 , target data for evaluating abnormal risk of business data is selected from the data list in order of arrangement.

[0113] Figure 8 The interface diagram of the data list in one embodiment of the present application is shown. As shown in the figure, the data list includes multiple fields such as customer, text content, hit keyword, selection probability and keyword type. The data in the table is sequentially sorted from top to bottom according to the arrangement order determined in step S620, and the text content closer to the top indicates that it represents the abnormal risk involved in the related business data more. The customer involved field is the customer to which the text content belongs, that is, the data source of the business data. By adding hyperlink text to the customer involved field, the basic information content of the customer can be viewed, and the specific text occurrence scene can be more conveniently understood. The hit keyword field can enable the user to clearly understand what keyword the text is hit by, and the keyword type can understand the risk type involved in the text.

[0114] In one embodiment of the present application, after the target data for evaluating the abnormal risk of the business data is extracted from the matching data, an evaluation message for evaluating the abnormal risk of the business data can also be automatically generated according to the business demand. In the embodiment of the present application, in response to the selection operation of the keyword type, the keyword text corresponding to the selected keyword type is extracted from the target data; and the keyword text is filled into a preset message template to generate an evaluation message for evaluating the abnormal risk of the business data. Taking the automatic generation of the message as an example, when the reviewer selects the risk type of the case, the keyword of the same risk type at a high level can be selected according to the risk type, and the related description of the text is automatically generated, thereby accelerating the generation efficiency of the message.

[0115] In the data processing method provided by the embodiment of the present application, all text information can be quickly matched by using the keyword database, the risk of the text information can be more quickly identified, and the comprehensive coverage of the risk is ensured. In the process of artificial daily review, the speed of due diligence can be greatly improved, and relevant risks are not missed. In addition, the embodiment of the present application directly outputs the text image representing the main risk of the case, which can enable the reviewer to quickly analyze the nature of the case, and avoid the situation that part of the text cannot be identified due to too much amount of text information, thereby ensuring compliance to a certain extent.

[0116] Taking the application of the embodiment of the present application in due diligence as an example, the embodiment of the present application can not only quickly complete the extraction and analysis of the text information of due diligence, but also ensure the timeliness, risk concentration degree and risk size balance of the main text risk extracted.

[0117] The artificial exhaustive search retrieves text information, and a single one needs more than 10s, and in the case of a large amount of text content involved, only 10%-20% of the data is generally extracted for analysis. The embodiment of the application can improve the speed by more than 80%, and at the same time, it can quickly retrieve 100% of the data, and can more comprehensively cover the risk.

[0118] And the artificial analysis cannot accurately quantify the various dimensions of the text information, while the risk extracted by the embodiment of the application is more objective, and compared with the artificial exhaustive search, the embodiment of the application can effectively guarantee the timeliness of the extracted text information and the like.

[0119] The embodiment of the application can effectively improve the efficiency of the audit exhaustive search process, and at the same time, improve the risk coverage degree, and effectively guarantee the compliance of the due diligence.

[0120] It should be noted that although the steps of the method in the application are described in a specific order in the drawings, this does not require or imply that the steps must be performed in that specific order, or that all the steps shown must be performed to achieve the desired result. In addition or alternatively, some steps can be omitted, a plurality of steps can be combined into one step, and / or one step can be divided into a plurality of steps, etc.

[0121] The following describes the device embodiment of the application, which can be used to execute the data processing method in the above-mentioned embodiments of the application. Figure 9 The structure block diagram of the data processing device provided by the embodiment of the application is schematically shown. As shown in Figure 9 The data processing device 900 includes: a business data acquisition module 910 configured to acquire business data to be processed, the business data including business text and business attributes associated with the business text, the business attributes being used to indicate data collection sources and data collection times of the business data; a keyword data acquisition module 920 configured to acquire keyword data used for abnormality detection of the business data, the keyword data including keyword text and keyword attributes associated with the keyword text, the keyword attributes being used to indicate keyword types and abnormality risk levels of the keyword data; a matching detection module 930 configured to perform matching detection on the keyword text and the business text to determine whether the keyword text is contained in the business text; a data combination module 940 configured to combine the business data and the keyword data to form matching data if the keyword text and the business text match successfully; and a data extraction module 950 configured to extract target data used for evaluating abnormality risk of the business data from the matching data.

[0122] In some embodiments of the present application, based on the above technical solutions, the data extraction module 950 comprises: a quantization processing module configured to perform quantization processing on the matching data according to the business attribute and the keyword attribute, to obtain quantization data corresponding to the matching data, the quantization data being numerical data comprising one or more numerical fields; a data sorting module configured to sort the matching data according to the values of the numerical fields in the quantization data, to obtain a data list arranged in order; and a data selection module configured to select target data for evaluating the abnormal risk of the business data from the data list in the order of arrangement.

[0123] In some embodiments of the present application, based on the above technical solutions, the quantization processing module comprises: a field extraction module configured to extract a first business field related to the business attribute, a second business field related to the keyword attribute, and a third business field related to statistical information of the matching data from the matching data; a field quantization module configured to perform quantization processing on the first business field, the second business field, and the third business field respectively according to the field types of the respective business fields, to obtain numerical fields corresponding to the respective business fields; and a field combination module configured to combine the numerical fields to obtain the quantization data corresponding to the matching data.

[0124] In some embodiments of the present application, based on the above technical solutions, the first business field comprises a time field for indicating the collection time of the business data, the second business field comprises a level field for indicating the abnormal risk level of the keyword data, and the third business field comprises a frequency field for indicating the occurrence frequency of the keyword text in the matching data.

[0125] In some embodiments of the present application, based on the above technical solutions, the data sorting module comprises: a weight acquisition module configured to acquire quantization weights corresponding to the numerical fields in the quantization data; a field weighting module configured to perform weighted summation on the numerical fields according to the quantization weights, to obtain a selection probability of the matching data; and a probability sorting module configured to sort the matching data according to the selection probability.

[0126] In some embodiments of the present application, based on the above technical solutions, the weight acquisition module comprises: a sample acquisition module configured to acquire quantization data samples obtained by performing quantization processing on historical business data, and to acquire data labels obtained by performing abnormal risk evaluation on the historical business data; a sample combination module configured to combine the quantization data samples and the data labels to form training samples; and a weight prediction module configured to predict quantization weights corresponding to the numerical fields in the quantization data according to the training samples.

[0127] In some embodiments of the present application, based on the above technical solutions, the weight prediction module comprises: a model acquisition module configured to acquire a weight prediction model used for weight prediction on the quantized data; an iterative training module configured to perform iterative training on the weight prediction model according to the training samples; and a model prediction module configured to predict quantized weights corresponding to each numerical field in the quantized data by the trained weight prediction model.

[0128] In some embodiments of the present application, based on the above technical solutions, the weight prediction module comprises: a matrix combination module configured to respectively form data sample matrices and data label matrices from quantized data samples and data labels in the training samples according to data types; and a matrix operation module configured to perform matrix operations on the data sample matrices and the data label matrices to obtain a weight matrix, the weight matrix comprising quantized weights corresponding to each numerical field in the quantized data.

[0129] In some embodiments of the present application, based on the above technical solutions, the matrix combination module is configured to: perform screening processing on the training samples to obtain a set number of related samples having data correlation with the business data; and respectively form data sample matrices and data label matrices from quantized data samples and data labels in the related samples according to data types.

[0130] In some embodiments of the present application, based on the above technical solutions, the data sorting module is configured to: acquire sorting priorities corresponding to each numerical field in the quantized data; and sort the matching data according to the sorting priorities and the values of each numerical field.

[0131] In some embodiments of the present application, based on the above technical solutions, the data processing apparatus further comprises: a hyperlink adding module configured to add a hyperlink text related to the data source of the business data in the data list; and a cloud map display module configured to display a keyword cloud map related to the data source of the business data in response to a triggering operation on the hyperlink text.

[0132] In some embodiments of the present application, based on the above technical solutions, the data processing apparatus further comprises: a text selection module configured to extract keyword texts corresponding to a selected keyword type from the target data in response to a selection operation on the keyword type; and a message generation module configured to fill the keyword texts into a preset message template to generate an evaluation message for evaluating the abnormal risk of the business data.

[0133] The specific details of the data processing apparatus provided in the embodiments of the present application have been described in detail in the corresponding method embodiments, and will not be repeated here.

[0134] Figure 10 A computer system structure block diagram of an electronic device for implementing the embodiments of the present application is schematically shown.

[0135] It should be noted that, Figure 10 The computer system 1000 of the electronic device shown is only an example, and should not bring any limitation to the functions and use range of the embodiments of the present application.

[0136] As Figure 10 shown, the computer system 1000 includes a central processing unit 1001 (CPU) which can perform various appropriate actions and processes according to programs stored in a read-only memory 1002 (ROM) or loaded from a storage portion 1008 into a random access memory 1003 (RAM). Various programs and data required for system operation are also stored in the random access memory 1003. The central processing unit 1001, the read-only memory 1002, and the random access memory 1003 are connected to each other through a bus 1004. An input / output interface 1005 (I / O interface) is also connected to the bus 1004.

[0137] The following components are connected to the input / output interface 1005: an input portion 1006 including a keyboard, a mouse, and the like; an output portion 1007 including a cathode ray tube (CRT), a liquid crystal display (LCD), and the like, and a speaker, and the like; a storage portion 1008 including a hard disk, and the like; and a communication portion 1009 including a network interface card such as a local area network card, a modem, and the like. The communication portion 1009 performs communication processing via a network such as the Internet. A drive 1010 is also connected to the input / output interface 1005 as necessary. A removable recording medium 1011 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, and the like is attached to the drive 1010 as necessary, so that a computer program read therefrom is installed in the storage portion 1008 as necessary.

[0138] In particular, according to embodiments of the present application, the processes described in the various method flowcharts can be implemented as a computer software program. For example, embodiments of the present application include a computer program product comprising a computer program carried on a computer readable medium, the computer program containing program code for executing the methods illustrated by the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network by the communication section 1009, and / or installed from the removable media 1011. When the computer program is executed by the central processing unit 1001, various functions defined in the system of the present application are executed.

[0139] It should be noted that the computer readable medium shown in the embodiments of the present application can be a computer readable signal medium or a computer readable storage medium or any combination of the two. The computer readable storage medium may, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or apparatus, or any combination of the above. More specific examples of the computer readable storage medium can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (Compact Disc Read-Only Memory, CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, the computer readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device or apparatus. In the present application, the computer readable signal medium can include a data signal carried in a baseband or as a part of a carrier wave, which carries computer readable program code. Such a propagated data signal can take many forms, including but not limited to an electromagnetic signal, an optical signal or any suitable combination of the above. The computer readable signal medium can also be any computer readable medium other than the computer readable storage medium, which can send, propagate or transmit the program for use by or in conjunction with an instruction execution system, device or apparatus. The program code contained on the computer readable medium can be transmitted by any suitable medium, including but not limited to wireless, wired, or the like, or any suitable combination of the above.

[0140] The computer program product of the present application can be a storage medium or a computer program product storing computer instructions. This storage medium or computer program product can be realized in many ways, such as computer data storage means (floppy disk, CD-ROM, USB, mobile hard disk, etc.), program cartridges of programmable computers, etc. The program instructions can be stored in the computer program product in many ways, such as object code, for example. The software implemented according to the embodiments of the present application is stored in the storage medium or computer program product, which can be any entity or device (computer memory, programming of memory, optical storage device, etc.) containing, storing, communicating, providing or transporting the program for use by or in connection with the processor.

[0141] It should be noted that although several modules or units of the device for action execution are mentioned in the foregoing detailed description, such a division is not mandatory. Indeed, according to the embodiments of the present application, the features and functionalities of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functionalities of one module or unit described above can be further divided into embodied by several modules or units.

[0142] From the above description of the embodiments, those skilled in the art will readily appreciate that the example embodiments described herein can be implemented by software and / or by hardware coupled with software. Accordingly, the technical solutions according to the embodiments of the present application can be embodied in the form of a software product. The software product can be stored in a non-volatile storage medium (which can be a CD-ROM, U disk, mobile hard disk, etc.) or network, and includes a number of instructions to make a computing device (which can be a personal computer, server, touch terminal, or network device, etc.) execute the methods according to the embodiments of the present application.

[0143] Other embodiments of the present application will be apparent to those skilled in the art from consideration of the specification and practice of the application disclosed herein. It is intended that the present application cover any and all variations of the application that come within the scope of the present application, along with all of the equivalents thereof. It is intended that the specification and examples be considered as exemplary only, with the true scope and spirit of the application being indicated by the following claims.

[0144] It should be understood that the present application is not limited to the precise construction that has been described above and illustrated in the accompanying drawings, and that various modifications and changes can be made by those skilled in the art without departing from the scope of the present application. The scope of the present application is limited only by the appended claims.

Claims

1. A data processing method, characterized by, The method comprises the following steps: obtaining business data to be processed from a business information table, the business information table comprising a plurality of pieces of business data, the business data being transaction data obtained after screening and information collection on large transactions or suspicious transactions found in monitoring, the business data comprising business text and business attributes associated with the business text, the business attributes being used to represent data collection sources and data collection times of the business data; obtaining keyword data used for abnormality detection of the business data, the keyword data comprising keyword text and keyword attributes associated with the keyword text, the keyword attributes being used to represent keyword types and abnormality risk levels of the keyword data; the keyword types comprising normal text and abnormal text, the normal text being used to represent that the keyword text does not have abnormality risk, and the abnormal text being used to represent that the keyword text has abnormality risk; performing matching detection on the keyword text and business text extracted one by one from the business information table to determine whether the keyword text is contained in the business text; if the keyword text and the business text are successfully matched, combining the business data and the keyword data to form matching data, the matching data being a new field combination formed by arranging fields in a specified dimension order; quantitatively processing the matching data according to the business attributes and the keyword attributes to obtain quantized data corresponding to the matching data, the quantized data being numerical data comprising one or more numerical fields; obtaining quantized data samples obtained by quantitatively processing historical business data, and obtaining data labels obtained by evaluating abnormality risk of the historical business data; combining the quantized data samples and the data labels to form training samples; Quantitative data samples and data labels in the training samples are respectively grouped according to data types to form a data sample matrix and a data label matrix; matrix operations are performed on the data sample matrix and the data label matrix according to a formula to obtain a weight matrix, A for the weight matrix, X for the data sample matrix, Y for the data label matrix, the weight matrix comprising quantitative weights corresponding to each numerical field in the quantitative data. performing weighted summation on the numerical fields according to the quantization weights to obtain a selection probability of the matching data; sorting the matching data according to the selection probability to obtain a data list in an ordered arrangement; selecting target data for evaluating abnormality risk of the business data from the data list according to the arrangement order.

2. The data processing method of claim 1, wherein, The quantitatively processing the matching data according to the business attributes and the keyword attributes comprises: extracting a first business field related to the business attributes, a second business field related to the keyword attributes, and a third business field related to statistical information of the matching data from the matching data; quantitatively processing the first business field, the second business field, and the third business field according to field types of the respective business fields to obtain numerical fields corresponding to the respective business fields; combining the numerical fields to obtain quantized data corresponding to the matching data.

3. The data processing method according to claim 2, characterized in that, The first business field comprises a time field used to represent collection times of the business data, the second business field comprises a level field used to represent abnormality risk levels of the keyword data, and the third business field comprises a frequency field used to represent frequencies of the keyword text in the matching data.

4. The data processing method of claim 1, wherein, The method further comprises: obtaining a weight prediction model for predicting weights of the quantized data; training the weight prediction model iteratively according to the training samples; predicting quantized weights corresponding to each numerical field in the quantized data by the trained weight prediction model.

5. The data processing method of claim 1, wherein, The quantized data samples and data labels in the training samples are respectively grouped into a data sample matrix and a data label matrix according to data types, including: filtering the training samples to obtain a set number of relevant samples having data correlation with the business data; The quantized data samples and data labels in the relevant samples are respectively grouped into a data sample matrix and a data label matrix.

6. The data processing method of claim 1, wherein, According to the values of each numerical field in the quantized data, the matching data is sorted, including: obtaining a sorting priority corresponding to each numerical field in the quantized data; According to the sorting priority and the value of each numerical field, the matching data is sorted.

7. The data processing method of claim 1, wherein, After sorting the matching data according to the values of each numerical field in the quantized data to obtain an ordered data list, the method further includes: adding a hyperlink text related to the data source of the business data in the data list; In response to a triggering operation on the hyperlink text, a keyword cloud related to the data source of the business data is displayed.

8. The data processing method according to any one of claims 1 to 7, characterized in that, After extracting target data for evaluating the abnormal risk of the business data from the matching data, the method further includes: In response to a selection operation on a keyword type, extracting keyword text corresponding to the selected keyword type from the target data; filling the keyword text into a preset message template to generate an evaluation message for evaluating the abnormal risk of the business data.

9. A data processing apparatus, characterized by including: a business data acquisition module configured to acquire business data to be processed from a business information table, the business information table including a plurality of business data, the business data being transaction data obtained after screening and information collection of large transactions or suspicious transactions found in monitoring, the business data including business text and business attributes associated with the business text, the business attributes being used to indicate data collection source and data collection time of the business data; a keyword data acquisition module configured to acquire keyword data for detecting abnormalities of the business data, the keyword data including keyword text and keyword attributes associated with the keyword text, the keyword attributes being used to indicate keyword type and abnormal risk level of the keyword data; the keyword type includes normal text and abnormal text, the normal text being used to indicate that the keyword text has no abnormal risk, and the abnormal text being used to indicate that the keyword text has abnormal risk; a matching detection module configured to perform matching detection on the keyword text and business text extracted one by one from the business information table to determine whether the keyword text is contained in the business text; The data combination module is configured to combine the business data and the keyword data to form matching data if the keyword text matches the business text successfully, the matching data being a new field combination formed by arranging each field in a specified dimension order; The data extraction module is configured to quantitatively process the matching data according to the business attribute and the keyword attribute to obtain quantized data corresponding to the matching data, the quantized data being numerical data including one or more numerical fields; obtain quantification data samples obtained by quantifying historical business data, and obtain data labels obtained by performing abnormal risk evaluation on the historical business data; group the quantification data samples and the data labels into training samples; group quantification data samples and data labels in the training samples according to data types to form a data sample matrix and a data label matrix; perform matrix operation on the data sample matrix and the data label matrix according to a formula to obtain a weight matrix, A wherein the weight matrix, X the data sample matrix, Y the data label matrix, the weight matrix comprising quantification weights corresponding to each numerical field in the quantification data. The numerical fields are weighted and summed according to the quantization weight to obtain a selection probability of the matching data; the matching data is sorted according to the selection probability to obtain a data list arranged in order; and target data for evaluating the abnormal risk of the business data is selected from the data list according to the arrangement order.

10. A computer readable medium having stored thereon a computer program which, when executed by a processor, implements the data processing method of any one of claims 1 to 8.

11. An electronic device, comprising: comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to implement the data processing method of any one of claims 1 to 7 via execution of the executable instructions.

12. A computer program product comprising computer instructions, characterized in that, The computer instructions, when executed by a processor, implement the data processing method of any one of claims 1 to 8.

Images