Method for constructing risk assessment engine by using key threshold rule information technology

Abstract

The invention discloses a method and a system for constructing a risk assessment engine by using a key threshold rule information technology, a storage medium and electronic equipment. The method comprises the following steps: completing cargo information synchronization in database copying and event queues and generating parallel processing events; executing a mode matching engine and an address verification tool in the memory database to realize data verification; free text processing is completed in the data analysis engine and the real-time analysis engine, and a comprehensive evaluation result is output. According to the method, data real-time synchronization of a service system and a risk assessment engine is achieved through database replication and an event queue, a compensation mechanism is automatically triggered when replication fails or delays, it is ensured that a core risk event can be preferentially processed in a high-concurrency environment in combination with a priority queue and flow control logic, and the risk assessment efficiency is improved. Therefore, the real-time performance and the reliability of data transmission are improved.

Description

Technical Field

[0001] This invention relates to the field of risk assessment engine construction and large-scale data processing, and in particular to a method and system for constructing a risk assessment engine using key threshold rule information technology, as well as storage media and electronic devices. Background Technology

[0002] With the rapid development of international trade and cross-border logistics, the quantity and variety of import and export goods have exploded, significantly increasing the scale and complexity of data processing. To cope with the massive data flow, related fields have gradually introduced automated risk assessment engines to support customs clearance and supervision. Existing assessment methods mainly rely on rule bases, blacklists, and statistical analysis tools, combined with database storage and retrieval technologies to automate the screening and risk comparison of certain goods. Simultaneously, artificial intelligence text analysis and machine learning models are increasingly being applied to trade data processing to improve the accuracy and efficiency of risk identification. These developments provide the foundation for building larger-scale and more complex data risk assessment systems.

[0003] However, existing technologies still have significant shortcomings. On the one hand, most systems can only perform batch processing in traditional relational databases, lacking the high-throughput parallel processing capabilities of in-memory databases, making it difficult to cope with the high concurrency demands of real-time data streams. On the other hand, existing risk screening largely remains at the static rule matching level, unable to achieve rapid response and dynamic expansion under an event-driven architecture, resulting in difficulty in timely identification of high-risk goods in complex cross-border business scenarios. Furthermore, existing systems have limited capabilities in integrating multi-source heterogeneous data, lacking unified data interfaces and standardized processing procedures, leading to severe information silos and insufficient reliability of risk assessment results. Moreover, for the intelligent processing of free text descriptions, most methods rely solely on keyword retrieval, failing to achieve semantic classification and encoding based on deep learning, thus limiting the accurate capture of potential risk characteristics. Summary of the Invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by this invention is that existing risk assessment methods have insufficient processing performance to meet high concurrency requirements, lack of event-driven dynamic expansion capability due to single rule matching, and low reliability of results due to insufficient data standardization and multi-source integration. The invention also addresses the problem of how to achieve multi-dimensional risk analysis and output traceable comprehensive assessment results under in-memory database and parallel processing architecture.

[0006] To address the aforementioned technical problems, this invention provides the following technical solution: a method for constructing a risk assessment engine using key threshold rule information technology, comprising synchronizing cargo information and generating parallel processing events in database replication and event queues; performing data verification using a pattern matching engine and address verification tool in an in-memory database; and completing free text processing and outputting comprehensive assessment results in a data analysis engine and a real-time analysis engine.

[0007] Optionally, in the above-described method embodiments of the present invention, the database replication includes: selecting the goods information table to be verified from the business system relational database, and generating a replication transaction in real time when the data is written to the business system through a data synchronization mechanism, transmitting the replication transaction to the relational database of the risk assessment engine, the replication process adopting a log-triggered method, mapping the fields of the source table and the target table one by one, and comparing the hash values ​​of the source data and the target data through an error detection algorithm; when replication fails or the delay exceeds a preset threshold, the system triggers a compensation mechanism to automatically retry and generates an error log to enter the audit module, forming a traceable link; at the same time, the replicated goods information will generate an initial event identifier according to the table field rules, triggering the event queue to enter the pending state.

[0008] Optionally, in the above-described method embodiments of the present invention, the event queue includes, upon receiving a cargo information event generated by database replication, allocating the cargo information event to different queue branches. Each branch corresponds to a specific transportation mode and business scenario, including road import, road export, air import, air export, sea import, and sea export, and supports time-based subdivision conditions. When the departure time of road export cargo is less than or equal to 30 minutes, the cargo information event is added to a priority queue. The queue is implemented using an in-memory message middleware for parallel multi-threaded consumption. When queue congestion exceeds a set capacity, the system triggers flow control logic to delay or discard low-priority tasks and sends the discard records to the monitoring module, prioritizing the handling of core risk events.

[0009] Optionally, in the above-described method embodiments of the present invention, the in-memory database includes storing copied cargo information data and a predefined risk profile table. Upon receiving a trigger from an event queue, the specified cargo information is loaded into the in-memory database node for high-speed retrieval. The in-memory database employs distributed sharding technology, with each shard containing a cargo number index, a risk tag index, and a timestamp index. When the node capacity exceeds a threshold, the system adds new nodes through an automatic expansion mechanism and redistributes existing shards to the new nodes. When processing risk assessment tasks, the system uses the data-supported pattern matching engine and address verification tool provided by the in-memory database to call cached data in parallel.

[0010] Optionally, in the above-described method embodiments of the present invention, the pattern matching engine and address verification tool include: comparing the cargo information with the risk item list to determine whether the cargo order number is on the blacklist and whether the cargo category belongs to the high-risk control catalog; and verifying the matching of cargo weight, batch number and external high-risk item database through multi-source data integration; during the matching process, a hash mapping algorithm is used to quickly compare string fields, and the consistency between the order and the waybill is verified through the three-way matching logic; if the amount exceeds the limit or the recipient's identity is inconsistent, the system triggers an abnormal event and records it to the audit log; the address verification tool performs formatting, standardization, spelling correction and geocoding on the cargo address, compares the result with the standard address database, and prompts for correction and generates an abnormal mark if information is missing; finally, the pattern matching result and the address verification result are written together into the in-memory database and used as input for the data analysis engine.

[0011] Optionally, in the above-described method embodiments of the present invention, the data analysis engine includes: preprocessing, segmenting, stemming, and filtering stop words in the free text product description; generating text vector representations through word embedding or a pre-trained Transformer model; after completing feature engineering, training a classification model using a labeled training set, including Naive Bayes, Support Vector Machine, and Random Forest algorithms; during runtime, inputting the vectorized features of the product description into the model, outputting product category labels, and generating classification codes or combined codes according to predefined encoding rules, wherein the encoding rules include the form of category prefix plus serial number; the generated codes and risk scores are written together into an in-memory database for use by the real-time analysis engine; simultaneously, a graph database stores and visualizes the association between products and risk items for manual investigation and backtracking.

[0012] Optionally, in the above-described method embodiments of the present invention, the real-time analysis engine includes: after receiving the results from the pattern matching engine, address verification tool, and text analysis engine, temporarily storing the results in a memory data grid, and integrating them to generate a comprehensive code when the event trigger is activated; when the comprehensive code score exceeds a threshold, the system outputs a high-risk code and generates an interception instruction, the decision result is stored in the business database, and a fast retrieval interface is provided through the key-value database for regulatory personnel to call; at the same time, the audit module records the input and output of each risk assessment.

[0013] The beneficial effects of this invention are as follows: The method for constructing a risk assessment engine using key threshold rule information technology provided by this invention achieves real-time data synchronization between the business system and the risk assessment engine through database replication and event queues. It automatically triggers a compensation mechanism when replication fails or is delayed. Combined with priority queues and flow control logic, it ensures that core risk events are processed first in high-concurrency environments, thereby improving the real-time performance and reliability of data transmission. Furthermore, by executing a pattern matching engine and address verification tool in an in-memory database, parallel computing based on distributed sharding and high-speed retrieval is achieved. This enables rapid consistency comparison between cargo characteristics and risk lists, and ensures data accuracy through address standardization and geocoding, thereby improving the identification rate of abnormal goods and data credibility. Based on this, the data analysis engine and real-time analysis engine work together to perform semantic modeling and classification encoding of free text commodity descriptions. Combined with pattern matching and address verification results, multi-dimensional fusion is performed to ultimately generate a comprehensive risk code and drive interception or release instructions, while establishing a traceable decision-making chain. Therefore, this invention achieves high-performance real-time processing, accurate data verification, and intelligent decision output, solving the shortcomings of existing technologies in terms of throughput, accuracy, and scalability, and significantly improving the intelligence level of customs and cross-border trade supervision.

[0014] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0015] The above and other objects, features, and advantages of the present invention will become more apparent from the more detailed description of the embodiments of the invention in conjunction with the accompanying drawings. The drawings are provided to further illustrate the embodiments of the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings, the same reference numerals generally represent the same parts or steps.

[0016] Figure 1 This is a flowchart illustrating a method provided in an exemplary embodiment of the present invention.

[0017] Figure 2 This is a schematic diagram of the system provided in an exemplary embodiment of the present invention. Detailed Implementation

[0018] Hereinafter, exemplary embodiments according to the present invention will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of the present invention, and not all embodiments of the present invention. It should be understood that the present invention is not limited to the exemplary embodiments described herein.

[0019] It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of the invention.

[0020] Those skilled in the art will understand that the terms "first," "second," etc., in the embodiments of the present invention are only used to distinguish different steps, devices, or modules, and do not represent any specific technical meaning, nor do they indicate a necessary logical order between them.

[0021] It should also be understood that in the embodiments of the present invention, "multiple" can refer to two or more, and "at least one" can refer to one, two or more.

[0022] It should also be understood that any component, data or structure mentioned in the embodiments of the present invention can generally be understood as one or more unless explicitly defined or given contrary instructions in the context.

[0023] refer to Figure 1 As an embodiment of the present invention, a method for constructing a risk assessment engine using key threshold rule information technology is provided, comprising:

[0024] Step 101: Synchronize cargo information and generate parallel processing events in the database replication and event queue.

[0025] Furthermore, database replication includes selecting the goods information table to be verified from the business system's relational database, and generating a replication transaction in real time when the data is written to the business system through a data synchronization mechanism. The replication transaction is then transmitted to the relational database of the risk assessment engine. The replication process is log-triggered, with each field of the source and target tables mapped one by one. At the same time, an error detection algorithm compares the hash values ​​of the source and target data. If replication fails or the delay exceeds a preset threshold, the system triggers a compensation mechanism to automatically retry and generates an error log that enters the audit module, forming a traceable link. Simultaneously, the replicated goods information generates an initial event identifier according to the table field rules, triggering the event queue to enter a pending state.

[0026] It should be noted that the event queue includes, upon receiving cargo information events generated by database replication, allocating the cargo information events to different queue branches. Each branch corresponds to a specific transportation mode and business scenario, including road import, road export, air import, air export, sea import, and sea export. It also supports time-based subdivision conditions. When the departure time of road export cargo is less than or equal to 30 minutes, the cargo information event is added to the priority queue. The queue is implemented using an in-memory message middleware for parallel multi-threaded consumption. When the queue congestion exceeds the set capacity, the system triggers flow control logic to delay or discard low-priority tasks and sends the discard records to the monitoring module, prioritizing the handling of core risk events.

[0027] It should also be noted that the database replication process enables real-time data synchronization between the business system and the risk assessment engine. Specifically, the replication mechanism uses a log-triggered approach to ensure a one-to-one correspondence between the source and target table fields, and guarantees data integrity through hash verification. A compensation mechanism is also introduced; when replication fails or delays exceed a threshold, the system automatically retryes and generates traceable logs. Subsequently, the replicated data triggers an event queue, allocating cargo information events to queue branches corresponding to the business scenario and supporting time-based priority sorting, such as triggering rapid processing within 30 minutes before the departure of export cargo. This mechanism ensures that the risk assessment engine always obtains the latest cargo information, preventing risk assessment failures due to data delays. Simultaneously, the event queue can classify and process various transportation methods and business scenarios, and, combined with flow control logic, implement priority scheduling to ensure that critical risk events are processed first.

[0028] Step 102: Execute the pattern matching engine and address verification tool in the in-memory database to perform data verification.

[0029] Furthermore, the in-memory database includes stored copied cargo information data and a predefined risk profile table. Upon receiving a trigger from the event queue, it loads the specified cargo information into the in-memory database node and performs a high-speed retrieval operation. The in-memory database adopts distributed sharding technology, with each shard containing a cargo number index, a risk tag index, and a timestamp index. When the node capacity exceeds the threshold, the system adds new nodes through an automatic expansion mechanism and redistributes the existing shards to the new nodes. When processing risk assessment tasks, the system uses the data-supported pattern matching engine and address verification tool provided by the in-memory database to call cached data in parallel.

[0030] It should be noted that the pattern matching engine and address verification tool include: comparing the cargo information with the risk item list to determine whether the shipment number is on the blacklist and whether the cargo category belongs to the high-risk control catalog; and verifying the matching of cargo weight, batch number and external high-risk item database through multi-source data integration. During the matching process, a hash mapping algorithm is used to quickly compare string fields, and the consistency between the order and the waybill is verified through a three-way matching logic. If the amount exceeds the limit or the recipient's identity does not match, the system triggers an abnormal event and records it in the audit log. The address verification tool performs formatting, standardization, spell correction and geocoding on the cargo address, compares the results with the standard address database, and prompts for correction and generates an anomaly mark if information is missing. Finally, the pattern matching result and the address verification result are written together into the in-memory database and used as input for the data analysis engine.

[0031] It should also be noted that high-speed data retrieval and parallel processing are achieved by loading cargo information and risk profile tables into an in-memory database. This database employs distributed sharding and automatic expansion mechanisms, dynamically adding nodes when capacity limits are exceeded, ensuring millisecond-level response times even under high concurrency. The pattern matching engine utilizes a consistency comparison mechanism and multi-source data integration to quickly compare key fields such as cargo number, category, and weight with the risk list, and combines this with three-way matching logic to verify the consistency between orders and waybills, thereby accurately identifying dishonest goods or abnormal transactions. Simultaneously, the address verification tool formats, corrects spellings, and geocodes cargo addresses, comparing them with a standard address database to ensure the authenticity and standardization of address information. This mechanism addresses the problems of traditional risk comparison methods relying on batch processing, being slow, and lacking accuracy. Through parallel calls and high-speed matching supported by the in-memory database, the system can not only complete risk screening in seconds but also improve data quality by correcting address information, preventing false or erroneous addresses from masking potential risks.

[0032] Step 103: Complete the free text processing in the data analysis engine and the real-time analysis engine and output the comprehensive evaluation results.

[0033] Furthermore, the data analysis engine includes preprocessing, word segmentation, stemming, and stop word filtering of free text product descriptions, and generating text vector representations through word embedding or pre-trained Transformer models. After feature engineering, a classification model is trained using a labeled training set, including Naive Bayes, Support Vector Machine, and Random Forest algorithms. During runtime, the vectorized features of the product description are input into the model, which outputs product category labels and generates classification codes or combined codes according to predefined coding rules, including the form of category prefix plus serial number. The generated codes and risk scores are written to an in-memory database for use by the real-time analysis engine. Simultaneously, a graph database stores and visualizes the relationships between products and risk items for manual investigation and backtracking.

[0034] It should be noted that the real-time analysis engine includes receiving the results from the pattern matching engine, address verification tool, and text analysis engine, temporarily storing the results in an in-memory data grid, and integrating them to generate a comprehensive code when an event trigger is activated. When the comprehensive code score exceeds a threshold, the system outputs a high-risk code and generates an interception instruction. The decision result is stored in the business database and a fast retrieval interface is provided through a key-value database for regulatory personnel to use. At the same time, the audit module records the input and output of each risk assessment.

[0035] It should also be noted that, through the data analysis engine, the system can perform preprocessing, word segmentation, stemming, and stop word filtering on free text product descriptions, and convert the text into vectorized representations using deep learning methods (such as word embeddings and Transformer models). Combined with trained classification models (Naive Bayes, SVM, Random Forest, etc.), it achieves automatic identification and encoding of product categories, ensuring standardized classification of complex and diverse product information. Simultaneously, the graph database stores the relationships between products and risk items, providing an intuitive visualization tool for manual backtracking and investigation. Based on this, the real-time analysis engine summarizes the results of pattern matching, address verification, and text analysis, integrating information in an in-memory data grid using an event-driven mechanism, and generating comprehensive codes based on scoring rules. When the score exceeds a threshold, the system automatically generates an interception command, and the results are synchronously written to the business database and quickly queried through a key-value database. The audit module fully records input and output, ensuring end-to-end traceability. This mechanism addresses the challenge of handling the fusion of free text and multi-source analysis results using existing technologies. Through deep semantic understanding and scoring mechanisms, the system can arrive at more accurate comprehensive risk assessment results based on multi-dimensional information, and directly drive decisions on interception, review or release.

[0036] Reference Figure 2 As an embodiment of the present invention, a system for constructing a risk assessment engine using key threshold rule information technology is provided, including a parallel processing module 21, a data verification module 22, and a comprehensive evaluation module 23.

[0037] The parallel processing module 21 is used to synchronize cargo information and generate parallel processing events in the database replication and event queue; the data verification module 22 is used to perform data verification by executing the pattern matching engine and address verification tool in the memory database; and the comprehensive evaluation module 23 is used to complete free text processing and output comprehensive evaluation results in the data analysis engine and real-time analysis engine.

[0038] The basic principles of this disclosure have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this disclosure are merely examples and not limitations, and should not be considered as essential features of each embodiment of this disclosure. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the scope of this disclosure to the necessity of employing the aforementioned specific details for implementation.

[0039] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For system embodiments, since they largely correspond to method embodiments, the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0040] The block diagrams of devices, apparatuses, devices, and systems disclosed herein are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.

[0041] The methods and apparatus of this disclosure may be implemented in many ways. For example, they may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above-described order of steps for the methods is for illustrative purposes only, and the steps of the methods of this disclosure are not limited to the order specifically described above unless otherwise specifically stated. Furthermore, in some embodiments, this disclosure may also be implemented as a program recorded on a recording medium, the program including machine-readable instructions for implementing the methods according to this disclosure. Thus, this disclosure also covers recording media storing programs for performing the methods according to this disclosure.

[0042] It should also be noted that in the apparatus, devices, and methods of this disclosure, the components or steps are decomposable and / or recombinable. Such decomposition and / or recombination should be considered equivalent to the present disclosure. The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein can be applied to other aspects without departing from the scope of this disclosure. Therefore, this disclosure is not intended to be limited to the aspects shown herein, but rather to be carried out within the widest scope consistent with the principles and novel features disclosed herein.

[0043] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of this disclosure to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations therein.

Claims

1. A method of constructing a risk assessment engine with critical threshold rules information technology, characterized by, Comprise: Synchronize goods information in database replication and event queue, and generate parallel processing events; Execute pattern matching engine and address verification tool in memory database to implement data verification; Complete free text processing in data analysis engine and real-time analysis engine, and output comprehensive evaluation results.

2. The method of building a risk assessment engine with critical threshold rules information technology as claimed in claim 1, wherein: The database replication comprises, Select the goods information table that needs to be verified from the business system relational database, and generate replication transactions in real time when data is written to the business system through the data synchronization mechanism, and transmit the replication transactions to the relational database of the risk assessment engine, the replication process adopts log trigger mode, the fields of the source table and the target table are mapped one by one, and the hash values of the source data and the target data are compared through the error detection algorithm; When the replication fails or the delay exceeds the preset threshold, the system triggers the compensation mechanism to automatically retry, and generates error logs to enter the audit module to form a traceable link; At the same time, the replicated goods information will generate an initial event identifier according to the table field rules, triggering the event queue to enter the processing state.

3. The method of building a risk assessment engine with critical threshold rules information technology as claimed in claim 2, wherein: The event queue comprises, After receiving the goods information event generated by the database replication, the goods information event is distributed to different queue branches, each branch corresponds to a specific transportation mode and business scenario, including highway import, highway export, air import, air export, sea import, sea export, and supports time subdivision conditions, when the highway export goods departure time is less than or equal to 30 minutes, the goods information event is added to the priority queue; The queue is implemented by using an in-memory message middleware, and is consumed in parallel multi-thread; When the queue congestion exceeds the set capacity, the system triggers the flow control logic to delay or discard low-priority tasks, and sends the discarded records to the monitoring module to prioritize core risk events.

4. The method of building a risk assessment engine with critical threshold rule information technology as claimed in claim 3, wherein: The memory database comprises, Store the replicated goods information data and the pre-defined risk profile table, and load the specified goods information to the memory database node after receiving the trigger of the event queue to perform high-speed retrieval operation; The memory database adopts distributed sharding technology, each shard contains goods number index, risk label index and timestamp index; When the node capacity exceeds the threshold, the system increases new nodes through the automatic expansion mechanism, and redistributes the existing shards to the new nodes; When processing risk assessment tasks, the system provides pattern matching engine and address verification tool through the data support mode of the memory database, and calls the cached data in parallel.

5. The method of building a risk assessment engine with key threshold rule information technology as claimed in claim 4, wherein: The pattern matching engine and address verification tool comprise, Compare the goods information with the risk item list for consistency, judge whether the goods number is in the credit blacklist, whether the goods category belongs to the high-risk regulated directory, and verify the matching of goods weight, batch number and external high-risk product library through multi-source data integration; In the matching process, the hash mapping algorithm is used to quickly compare the string fields, and the three single collision logic is used to verify the consistency of the order and the waybill, if the amount exceeds the limit or the recipient's identity is inconsistent, the system triggers an exception event and records it to the audit log; The address verification tool performs formatting, standardization, spelling correction and geocoding on the cargo address, compares the result with a standard address library, and prompts for correction and generates an exception label if information is missing; the final pattern matching result and the address verification result are written into the in-memory database and used as input for the data analysis engine.

6. The method of building a risk assessment engine with key threshold rule information technology as claimed in claim 5, wherein: The data analysis engine includes, The free text commodity description is preprocessed, tokenized, stemmed, and stop word filtered, and a text vector representation is generated using word embedding or a pre-trained Transformer model; After feature engineering is completed, the labeled training set is used to train the classification model, including naive Bayes, support vector machine and random forest algorithm; During runtime, the vectorized features of the commodity description are input into the model, the commodity category label is output, and the classification code or combined code is generated according to the predefined coding rule, wherein the coding rule includes the form of category prefix plus serial number; The generated code and risk score are written into the in-memory database for real-time analysis engine to call, and the graph database stores and visualizes the association between the commodity and the risk item for manual investigation and backtracking.

7. The method of building a risk assessment engine with key threshold rule information technology as claimed in claim 6, wherein: The real-time analysis engine includes, After receiving the results of the pattern matching engine, the address verification tool and the text analysis engine, the results are temporarily stored in the in-memory data grid, and are integrated to generate a comprehensive code when the event trigger is triggered; When the comprehensive code score exceeds the threshold, the system outputs a high-risk code and generates an interception instruction, the decision result is stored in the business database, and a fast retrieval interface is provided through the key-value database for supervisors to call, and the audit module records the input and output of each risk assessment.

8. A system for building a risk assessment engine with critical threshold rules information technology, employing the method of any one of claims 1 to 7, characterized in that: It includes a parallel processing module, a data verification module and a comprehensive evaluation module; The parallel processing module is used to synchronize the cargo information in the database replication and the event queue and generate a parallel processing event; The data verification module is used to perform data verification in the in-memory database using the pattern matching engine and the address verification tool; The comprehensive evaluation module is used to complete free text processing in the data analysis engine and the real-time analysis engine and output a comprehensive evaluation result.

9. An electronic device, comprising: The electronic device includes a memory and a processor coupled to the memory; the memory stores program instructions, and the program instructions are executed by the processor to cause the electronic device to perform the method of any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer program, when executed on an electronic device, causes the electronic device to perform the method of any one of claims 1-7.

Images