Trade background authenticity confirmation method and device, computer equipment and storage medium

Abstract

The invention relates to a trade background authenticity confirmation method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring enterprise trade data; identifying entities in the enterprise trade data, entity types to which the entities belong and semantic relationships among the entities; a trade event is extracted from the enterprise trade data; based on the entity, the entity type, the semantic relationship and the trade event, constructing a trade knowledge graph; performing quantitative evaluation on the authenticity of the trade background from multiple dimensions based on the trade knowledge graph to obtain an evaluation score of each dimension; and determining a trade background authenticity level based on the evaluation score. By adopting the method, the authenticity of the trade background can be confirmed by deeply fusing semantic understanding and big data.

Description

Technical Field

[0001] This application relates to the field of natural language processing technology, and in particular to a method, apparatus, computer equipment, computer-readable storage medium, and computer program product for verifying the authenticity of trade background. Background Technology

[0002] As a core component of supply chain risk control, the assessment of the authenticity of trade background is directly related to the credit security and business risks of financial institutions.

[0003] Traditional methods for assessing the authenticity of trade backgrounds primarily rely on manual review and rule engines. However, over-reliance on manual review is not only inefficient but also prone to inconsistencies due to differences in reviewer experience or subjective judgment. While rule engines can automate certain processes, they struggle to flexibly handle new or complex trade models, making them susceptible to misjudgments or omissions. Furthermore, because trade chains involve multiple stakeholders and stages, information is often scattered across different systems and institutions, leading to significant information fragmentation and hindering the formation of a comprehensive perspective for authenticity assessment.

[0004] In recent years, although methods applying artificial intelligence technology to trade background assessment have emerged, they still fall short in terms of deep semantic understanding and multi-dimensional comprehensive analysis. Therefore, there is an urgent need for a trade background authenticity assessment method that can deeply integrate semantic understanding and big data. Summary of the Invention

[0005] Therefore, it is necessary to provide a method, apparatus, computer equipment, computer-readable storage medium, and computer program product that can deeply integrate semantic understanding and big data to verify the authenticity of trade background in order to address the above-mentioned technical problems.

[0006] Firstly, this application provides a method for verifying the authenticity of trade background, including:

[0007] Obtain enterprise trade data;

[0008] Identify entities, entity types, and semantic relationships between entities in enterprise trade data; extract trade events from enterprise trade data; and construct a trade knowledge graph based on entities, entity types, semantic relationships, and trade events.

[0009] The authenticity of trade background is quantitatively assessed from multiple dimensions based on a trade knowledge graph, and an assessment score is obtained for each dimension; the level of authenticity of trade background is determined based on the assessment score.

[0010] In one embodiment, acquiring enterprise trade data includes:

[0011] Raw trade data was collected from multiple data sources;

[0012] Data preprocessing is performed on the raw trade data to obtain enterprise trade data. Data preprocessing includes: word segmentation, part-of-speech tagging and entity naming of unstructured text data in the raw trade data, format standardization and outlier removal of structured data in the raw trade data, and association of the same entity from different data sources.

[0013] In one embodiment, entities in enterprise trade data, the entity types to which the entities belong, and the semantic relationships between entities are identified; trade events are extracted from the enterprise trade data, including:

[0014] The pre-trained entity recognition model identifies entities in enterprise trade data and the entity types to which they belong.

[0015] Dependency parsing and semantic role labeling techniques are used to extract semantic relationships between entities;

[0016] Identify trade events from enterprise trade data using an event extraction model.

[0017] In one embodiment, multiple dimensions include entity authenticity, relationship rationality, event consistency, and behavioral compliance.

[0018] Based on a trade knowledge graph, the authenticity of trade background is quantitatively assessed from multiple dimensions, including:

[0019] The authenticity of entities in the trade knowledge graph is evaluated by comparing multi-source data, and an entity authenticity evaluation score is obtained.

[0020] Based on historical data and industry rules, the rationality of semantic relationships in the trade knowledge graph is determined, and a relationship rationality assessment score is obtained.

[0021] The consistency of trade events in the knowledge graph in terms of time sequence and logical order is examined to obtain an event consistency evaluation score.

[0022] The compliance of trade events in the knowledge graph is assessed, and a compliance assessment score is obtained.

[0023] In one embodiment, determining the level of authenticity of the trade background based on the assessment score includes:

[0024] Obtain the predefined weights corresponding to each dimension;

[0025] The overall authenticity score is determined based on predefined weights, entity authenticity assessment score, relationship rationality assessment score, event consistency assessment score, and compliance assessment score.

[0026] The overall authenticity score is compared with a preset threshold, and the authenticity level of the trade background is determined based on the comparison results.

[0027] In one embodiment, the method further includes:

[0028] After obtaining the evaluation score for each dimension, if the evaluation score is less than the set value, an evaluation description text and an improvement suggestion text are generated. Based on the evaluation description text and the improvement suggestion text, a risk assessment report is generated.

[0029] Secondly, this application also provides a device for verifying the authenticity of trade background, comprising:

[0030] The acquisition module is used to acquire enterprise trade data;

[0031] The identification module is used to identify entities, entity types, and semantic relationships between entities in enterprise trade data; extract trade events from enterprise trade data; and construct a trade knowledge graph based on entities, entity types, semantic relationships, and trade events.

[0032] The assessment module is used to quantitatively assess the authenticity of trade background from multiple dimensions based on the trade knowledge graph, and obtain an assessment score for each dimension; the level of authenticity of trade background is determined based on the assessment score.

[0033] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0034] Obtain enterprise trade data;

[0035] Identify entities, entity types, and semantic relationships between entities in enterprise trade data; extract trade events from enterprise trade data; and construct a trade knowledge graph based on entities, entity types, semantic relationships, and trade events.

[0036] The authenticity of trade background is quantitatively assessed from multiple dimensions based on a trade knowledge graph, and an assessment score is obtained for each dimension; the level of authenticity of trade background is determined based on the assessment score.

[0037] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0038] Obtain enterprise trade data;

[0039] Identify entities, entity types, and semantic relationships between entities in enterprise trade data; extract trade events from enterprise trade data; and construct a trade knowledge graph based on entities, entity types, semantic relationships, and trade events.

[0040] The authenticity of trade background is quantitatively assessed from multiple dimensions based on a trade knowledge graph, and an assessment score is obtained for each dimension; the level of authenticity of trade background is determined based on the assessment score.

[0041] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0042] Obtain enterprise trade data;

[0043] Identify entities, entity types, and semantic relationships between entities in enterprise trade data; extract trade events from enterprise trade data; and construct a trade knowledge graph based on entities, entity types, semantic relationships, and trade events.

[0044] The authenticity of trade background is quantitatively assessed from multiple dimensions based on a trade knowledge graph, and an assessment score is obtained for each dimension; the level of authenticity of trade background is determined based on the assessment score.

[0045] The aforementioned methods, apparatus, computer equipment, computer-readable storage media, and computer program products for verifying the authenticity of trade backgrounds effectively improve the accuracy and comprehensiveness of verifying the authenticity of trade backgrounds by acquiring enterprise trade data; identifying entities, entity types, and semantic relationships between entities in the enterprise trade data; extracting trade events from the enterprise trade data; constructing a trade knowledge graph based on entities, entity types, semantic relationships, and trade events; quantitatively evaluating the authenticity of trade backgrounds from multiple dimensions based on the trade knowledge graph to obtain an evaluation score for each dimension; and determining the level of authenticity of trade backgrounds based on the evaluation scores. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0047] Figure 1 This is a diagram illustrating the application environment of a trade background authenticity verification method in one embodiment.

[0048] Figure 2This is a flowchart illustrating a method for verifying the authenticity of trade background in one embodiment;

[0049] Figure 3 This is a flowchart illustrating the method for verifying the authenticity of trade background in another embodiment;

[0050] Figure 4 This is a structural block diagram of a trade background authenticity verification device in one embodiment;

[0051] Figure 5 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0052] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0053] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0054] The method for verifying the authenticity of trade background provided in this application can be applied to, for example... Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104 or placed on a cloud or other network server. Specifically, terminal 102 or server 104 completes a method for confirming the authenticity of trade background information. This method includes: acquiring enterprise trade data; identifying entities in the enterprise trade data, the entity types to which the entities belong, and the semantic relationships between entities; extracting trade events from the enterprise trade data; constructing a trade knowledge graph based on entities, entity types, semantic relationships, and trade events; quantitatively evaluating the authenticity of the trade background information from multiple dimensions based on the trade knowledge graph, obtaining an evaluation score for each dimension; and determining the level of authenticity of the trade background information based on the evaluation scores.

[0055] Terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, drones, low-altitude aircraft, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, and projection equipment. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted displays. Head-mounted displays can be virtual reality (VR) devices, augmented reality (AR) devices, and smart glasses. Server 104 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0056] In one exemplary embodiment, such as Figure 2 As shown, a method for verifying the authenticity of trade background is provided, which is then applied to... Figure 1 Taking terminal 102 as an example, the explanation includes the following steps 202 to 206. Wherein:

[0057] Step 202: Obtain enterprise trade data.

[0058] Among them, enterprise trade data is multi-source data that has been preprocessed. Multi-source data refers to data that can be collected from multiple data sources, such as customs data, enterprise data, third-party data, and public data.

[0059] Step 204: Identify entities, entity types, and semantic relationships in enterprise trade data; extract trade events from enterprise trade data; and construct a trade knowledge graph based on entities, entity types, semantic relationships, and trade events.

[0060] Among them, an entity is a participant, object, or concept involved in a company's trade activities; an entity type is a classification of entities with the same characteristics and attributes, such as: enterprises, products, people, or locations; a trade event is an event that occurs in trade activities, such as: signing a contract, paying a loan, or shipping goods; semantic relations define the relationships between entities and between entities of the same type, such as: supply relations, cooperation relations, and transaction relations; and a trade knowledge graph is an interconnected trade world model constructed by integrating entities, entity types, semantic relations, and trade events.

[0061] For example, entities, entity types, semantic relationships, and trade events are stored in a structured manner according to the trade domain ontology model to construct a trade knowledge graph.

[0062] Step 206: Quantitatively assess the authenticity of trade background from multiple dimensions based on the trade knowledge graph, and obtain the assessment score for each dimension; determine the level of authenticity of trade background based on the assessment score.

[0063] The authenticity of trade background is categorized into three levels: high credibility, medium credibility, and low credibility.

[0064] For example, predefined weights are obtained for each dimension; the evaluation scores of each dimension are weighted and summed based on the predefined weights to obtain a comprehensive authenticity score; and the authenticity level of the trade background is determined based on the comprehensive authenticity score.

[0065] The aforementioned method for verifying the authenticity of trade background involves acquiring enterprise trade data; identifying entities, entity types, and semantic relationships within the trade data; extracting trade events from the trade data; constructing a trade knowledge graph based on entities, entity types, semantic relationships, and trade events; quantitatively evaluating the authenticity of the trade background from multiple dimensions using the trade knowledge graph to obtain an evaluation score for each dimension; and determining the level of authenticity of the trade background based on the evaluation scores. This method effectively improves the accuracy and comprehensiveness of verifying the authenticity of trade background.

[0066] In one exemplary embodiment, acquiring enterprise trade data includes: collecting raw trade data from multiple data sources; performing data preprocessing on the raw trade data to obtain enterprise trade data; the data preprocessing includes: performing word segmentation, part-of-speech tagging, and entity naming on unstructured text data in the raw trade data, performing format standardization and outlier filtering on structured data in the raw trade data, and associating the same entity from different data sources.

[0067] Optionally, raw trade data can be data obtained from data systems, such as customs declarations, commodity lists, and tax bills from customs declaration systems, and purchase contracts, sales contracts, invoices, and logistics documents from enterprise resource planning systems. Raw trade data can also be data obtained from third-party data platforms, such as credit data, industry analysis reports, and market supervision information. Raw trade data can also be data obtained from publicly available data sources, such as enterprise registration information, equity structure, and business scope data.

[0068] For example, word segmentation is the process of dividing continuous natural language text in unstructured text data of raw trade data into multiple independent and meaningful words; part-of-speech tagging is the process of assigning specific grammatical category labels to multiple independent and meaningful words based on word segmentation; entity naming is the process of identifying entities with specific meanings from unstructured text in raw trade data and naming them.

[0069] Furthermore, format standardization refers to unifying the structured data in the original trade data into a consistent format, unit, and representation; outlier removal refers to identifying and removing extreme data points in the structured data of the original trade data that deviate significantly from other observations and may be caused by errors.

[0070] Optionally, ontology-based data fusion technology can be used to associate and unify the same entity from different data sources to obtain fused entity data; the integrity, consistency and accuracy of the fused entity data can be verified.

[0071] In this embodiment, by performing comprehensive and detailed preprocessing on the original trade data, multi-source heterogeneous data can be effectively integrated. This not only allows valuable information to be extracted from massive and complex enterprise trade data, providing strong support for accurately judging the authenticity of the trade background, but also lays a solid foundation for the subsequent construction of an accurate and complete trade knowledge graph.

[0072] In one embodiment, entities in enterprise trade data, the entity types to which the entities belong, and the semantic relationships between entities are identified; trade events are extracted from enterprise trade data, including: identifying entities in enterprise trade data and the entity types to which the entities belong based on a pre-trained entity recognition model; extracting semantic relationships between entities using dependency parsing and semantic role labeling techniques; and identifying trade events from enterprise trade data through an event extraction model.

[0073] Among them, the pre-trained entity recognition model is an entity recognition model pre-trained using a large-scale Chinese training corpus, used to identify entities in enterprise trade data and the entity types to which those entities belong. The event extraction model is a natural language processing model specifically designed to identify trade events from enterprise trade data. It can automatically identify events with specific meaning from unstructured plain text and structure them into a machine-readable and processable standard format.

[0074] For example, dependency parsing is used to display the grammatical dependencies between entities, such as subject-verb-object structures; semantic role labeling technology is used to reveal the participants in the action, i.e. entities. Combining dependency parsing and semantic role labeling technology can extract the semantic relationships between entities.

[0075] In this embodiment, by employing natural language processing models and technologies, entities, entity types, semantic relationships, and trade events in enterprise trade data can be accurately identified, ensuring the accuracy and completeness of the knowledge graph. This provides a reliable basis for subsequent quantitative assessment of the authenticity of trade background from multiple dimensions, effectively improving the reliability of the entire trade background authenticity confirmation process.

[0076] In one embodiment, multiple dimensions include entity authenticity, relationship rationality, event consistency, and behavioral compliance. The authenticity of trade background is quantitatively assessed from multiple dimensions based on a trade knowledge graph, including: evaluating the authenticity of entities in the trade knowledge graph by comparing multi-source data to obtain an entity authenticity assessment score; determining the rationality of semantic relationships in the trade knowledge graph based on historical data and industry rules to obtain a relationship rationality assessment score; verifying the consistency of trade events in the knowledge graph in terms of time sequence and logical order to obtain an event consistency assessment score; and assessing the compliance of trade events in the knowledge graph to obtain a compliance assessment score.

[0077] For example, the authenticity and consistency of entities in the trade knowledge graph are evaluated by comparing multi-source data to obtain an entity authenticity assessment score. For example, the authenticity of the supplier's business registration information is evaluated, whether the business scope matches the traded products is evaluated, and whether the credit record is good is evaluated. The entity authenticity assessment score is obtained by combining the assessment results.

[0078] For example, the reasonableness of semantic relationships in the trade knowledge graph is determined based on historical data and industry rules, and a reasonableness assessment score is obtained. For example, it is determined whether the transaction relationship between suppliers and buyers is stable, whether the transaction frequency and amount are in line with industry characteristics, and whether the transaction price is within a reasonable range. The reasonableness assessment score is obtained by combining the determination results.

[0079] For example, the consistency of trade events in the knowledge graph in terms of time sequence and logical order can be tested to obtain an event consistency assessment score. For example, testing whether the contract signing time is earlier than the delivery time, whether the payment time conforms to the contract agreement, and whether the logistics information matches the transaction record can be tested. The event consistency assessment score is obtained by combining the test results.

[0080] For example, the compliance of trade events in the knowledge graph is assessed to obtain a compliance assessment score. For example, the assessment may include whether the declared information complies with customs regulations, whether the transaction behavior violates market supervision requirements, and whether there are any abnormal transactions or fraud risks. The compliance assessment score is obtained by combining the assessment results.

[0081] In one embodiment, determining the authenticity level of the trade background based on the assessment score includes: obtaining predefined weights corresponding to each dimension; determining a comprehensive authenticity score based on the predefined weights, entity authenticity assessment score, relationship rationality assessment score, event consistency assessment score, and compliance assessment score; comparing the comprehensive authenticity score with a preset threshold, and determining the authenticity level of the trade background based on the comparison result.

[0082] The predefined weights can be adjusted according to specific industries and business scenarios.

[0083] For example, predefined weights are obtained for the entity authenticity dimension, relationship rationality dimension, event consistency dimension, and behavior compliance dimension; the entity authenticity score is obtained by multiplying the predefined weights of the entity authenticity dimension by the entity authenticity assessment score; the relationship rationality score is obtained by multiplying the predefined weights of the relationship rationality dimension by the relationship rationality assessment score; the event consistency score is obtained by multiplying the predefined weights of the event consistency dimension by the event consistency assessment score; the compliance score is obtained by multiplying the predefined weights of the compliance dimension by the compliance assessment score; the entity authenticity score, relationship rationality score, event consistency score, and compliance score are summed to obtain the overall authenticity score; the overall authenticity score is compared with preset thresholds for the overall authenticity scores corresponding to the three trade background authenticity levels of high credibility, medium credibility, and low credibility, and the trade background authenticity level is determined based on the comparison results.

[0084] In this embodiment, the authenticity of the trade background is quantitatively evaluated from multiple dimensions, and the comprehensive authenticity score is determined by combining the evaluation scores of each dimension and the predefined weights. Then, it is compared with the preset threshold to obtain the authenticity level of the trade background. This allows for the analysis of the trade background from different dimensions, so that the final determined authenticity level of the trade background can provide a reliable basis for trade decisions.

[0085] In one embodiment, the method further includes: after obtaining the evaluation score corresponding to each dimension, if the evaluation score is less than a set value, generating evaluation description text and improvement suggestion text, and generating a risk assessment report based on the evaluation description text and improvement suggestion text.

[0086] The assessment description text is used to identify potential risks, while the improvement recommendation text is used to propose countermeasures to address these potential risks.

[0087] Optionally, the assessment results can be presented to users intuitively through a visual dashboard, such as using a scoring radar chart, risk distribution chart, or trend analysis chart.

[0088] In this embodiment, by generating assessment description text and improvement suggestion text, and based on these texts generating a risk assessment report, enterprises can be provided with detailed risk analysis and improvement directions, providing strong data support and risk warning for their trade decisions.

[0089] Next reference Figure 3 The present application will be described in detail with a specific embodiment of the method for verifying the authenticity of the trade background.

[0090] Step 1: Collect relevant trade data from multiple data sources.

[0091] Step 2: Preprocess the collected data.

[0092] The system performs word segmentation, part-of-speech tagging, and named entity recognition on unstructured text data; it performs format standardization and outlier detection on structured data; it employs ontology-based data fusion technology to associate and unify the representation of the same entity from different sources; and it verifies the integrity, consistency, and accuracy of the fused data through a data quality assessment module.

[0093] Step 3: Perform semantic analysis on the preprocessed data and construct a knowledge graph.

[0094] A semantic understanding engine is used to perform deep semantic analysis on the preprocessed trade data, including entity recognition and classification, relation extraction, and event extraction. Finally, the extracted entities, attributes, and relations are structured and stored according to a trade domain ontology model to construct a trade knowledge graph.

[0095] Step 4: Conduct a multi-dimensional assessment based on the trade knowledge graph.

[0096] Based on the trade knowledge graph, the authenticity of entities, the rationality of relationships, the consistency of events, and compliance are evaluated from multiple dimensions. Each dimension is scored out of 100, and then a comprehensive score is calculated based on predefined weights.

[0097] Step 5: Result generation and display.

[0098] The weighted summation method is used to combine the assessment scores of the four dimensions into an overall authenticity score, and the trade background is divided into three levels: "high credibility", "medium credibility" and "low credibility" according to the risk threshold. Detailed assessment descriptions and improvement suggestions are generated for each assessment dimension, pointing out potential risk points and countermeasures. The assessment results are displayed to users intuitively through a visual dashboard.

[0099] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0100] Based on the same inventive concept, this application also provides a trade background authenticity verification device for implementing the trade background authenticity verification method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more embodiments of the trade background authenticity verification device provided below can be found in the limitations of the trade background authenticity verification method described above, and will not be repeated here.

[0101] In one exemplary embodiment, such as Figure 4 As shown, a trade background authenticity verification device 400 is provided, including: an acquisition module 402, an identification module 404, and an evaluation module 406, wherein:

[0102] Module 402 is used to acquire enterprise trade data.

[0103] The identification module 404 is used to identify entities in enterprise trade data, the entity types to which entities belong, and the semantic relationships between entities; extract trade events from enterprise trade data; and construct a trade knowledge graph based on entities, entity types, semantic relationships, and trade events.

[0104] The assessment module 406 is used to quantitatively assess the authenticity of trade background from multiple dimensions based on the trade knowledge graph, and obtain an assessment score for each dimension; the level of authenticity of trade background is determined based on the assessment score.

[0105] In one embodiment, the acquisition module is further configured to collect raw trade data from multiple data sources; perform data preprocessing on the raw trade data to obtain enterprise trade data; the data preprocessing includes: performing word segmentation, part-of-speech tagging and entity naming on unstructured text data in the raw trade data, performing format standardization and outlier filtering on structured data in the raw trade data, and associating the same entity from different data sources.

[0106] In one embodiment, the identification module is further configured to identify entities in enterprise trade data and the entity types to which the entities belong based on a pre-trained entity recognition model; extract semantic relationships between entities using dependency parsing and semantic role labeling techniques; and identify trade events from enterprise trade data using an event extraction model.

[0107] In one embodiment, multiple dimensions include entity authenticity, relationship rationality, event consistency, and behavioral compliance. The evaluation module is also used to evaluate the authenticity of entities in the trade knowledge graph by comparing multi-source data to obtain an entity authenticity evaluation score; determine the rationality of semantic relationships in the trade knowledge graph based on historical data and industry rules to obtain a relationship rationality evaluation score; examine the consistency of trade events in the knowledge graph in terms of time sequence and logical order to obtain an event consistency evaluation score; and evaluate the compliance of trade events in the knowledge graph to obtain a compliance evaluation score.

[0108] In one embodiment, the assessment module is further configured to obtain predefined weights corresponding to each dimension; determine a comprehensive authenticity score based on the predefined weights, entity authenticity assessment score, relationship rationality assessment score, event consistency assessment score, and compliance assessment score; compare the comprehensive authenticity score with a preset threshold, and determine the trade background authenticity level based on the comparison result.

[0109] In one embodiment, the evaluation module is further configured to, after obtaining the evaluation score corresponding to each dimension, generate evaluation description text and improvement suggestion text if the evaluation score is less than a set value, and generate a risk assessment report based on the evaluation description text and improvement suggestion text.

[0110] The modules in the aforementioned trade background authenticity verification device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0111] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 5As shown, the computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores enterprise trade data. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network. When executed by the processor, the computer program implements a method for verifying the authenticity of trade background information.

[0112] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0113] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0114] Obtain enterprise trade data;

[0115] Identify entities, entity types, and semantic relationships between entities in enterprise trade data; extract trade events from enterprise trade data; and construct a trade knowledge graph based on entities, entity types, semantic relationships, and trade events.

[0116] The authenticity of trade background is quantitatively assessed from multiple dimensions based on a trade knowledge graph, and an assessment score is obtained for each dimension; the level of authenticity of trade background is determined based on the assessment score.

[0117] In one embodiment, when the processor executes the computer program, it also performs the following steps: collecting raw trade data from multiple data sources; performing data preprocessing on the raw trade data to obtain enterprise trade data; the data preprocessing includes: segmenting, tagging, and naming unstructured text data in the raw trade data; standardizing the format and removing outliers from the structured data in the raw trade data; and associating the same entity from different data sources.

[0118] In one embodiment, when the processor executes the computer program, it also performs the following steps: identifying entities in enterprise trade data and the entity types to which the entities belong based on a pre-trained entity recognition model; extracting semantic relationships between entities using dependency parsing and semantic role labeling techniques; and identifying trade events from enterprise trade data using an event extraction model.

[0119] In one embodiment, when the processor executes the computer program, it further performs the following steps: assessing the authenticity of entities in the trade knowledge graph by comparing multi-source data to obtain an entity authenticity assessment score; determining the reasonableness of semantic relationships in the trade knowledge graph based on historical data and industry rules to obtain a relationship reasonableness assessment score; verifying the consistency of trade events in the knowledge graph in terms of time sequence and logical order to obtain an event consistency assessment score; and assessing the compliance of trade events in the knowledge graph to obtain a compliance assessment score.

[0120] In one embodiment, when the processor executes the computer program, it also performs the following steps: obtaining predefined weights corresponding to each dimension; determining a comprehensive authenticity score based on the predefined weights, entity authenticity assessment score, relationship rationality assessment score, event consistency assessment score, and compliance assessment score; comparing the comprehensive authenticity score with a preset threshold, and determining the trade background authenticity level based on the comparison result.

[0121] In one embodiment, when the processor executes the computer program, it further performs the following steps: after obtaining the evaluation score corresponding to each dimension, if the evaluation score is less than a set value, generating evaluation description text and improvement suggestion text, and generating a risk assessment report based on the evaluation description text and improvement suggestion text.

[0122] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0123] Obtain enterprise trade data;

[0124] Identify entities, entity types, and semantic relationships between entities in enterprise trade data; extract trade events from enterprise trade data; and construct a trade knowledge graph based on entities, entity types, semantic relationships, and trade events.

[0125] The authenticity of trade background is quantitatively assessed from multiple dimensions based on a trade knowledge graph, and an assessment score is obtained for each dimension; the level of authenticity of trade background is determined based on the assessment score.

[0126] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: collecting raw trade data from multiple data sources; preprocessing the raw trade data to obtain enterprise trade data; the data preprocessing includes: segmenting, tagging, and naming unstructured text data in the raw trade data; standardizing the format and removing outliers from the structured data in the raw trade data; and associating the same entity from different data sources.

[0127] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: identifying entities in enterprise trade data and the entity types to which the entities belong based on a pre-trained entity recognition model; extracting semantic relationships between entities using dependency parsing and semantic role labeling techniques; and identifying trade events from enterprise trade data using an event extraction model.

[0128] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: assessing the authenticity of entities in the trade knowledge graph by comparing multi-source data to obtain an entity authenticity assessment score; determining the reasonableness of semantic relationships in the trade knowledge graph based on historical data and industry rules to obtain a relationship reasonableness assessment score; verifying the consistency of trade events in the knowledge graph in terms of time sequence and logical order to obtain an event consistency assessment score; and assessing the compliance of trade events in the knowledge graph to obtain a compliance assessment score.

[0129] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: obtaining predefined weights corresponding to each dimension; determining a comprehensive authenticity score based on the predefined weights, entity authenticity assessment score, relationship rationality assessment score, event consistency assessment score, and compliance assessment score; comparing the comprehensive authenticity score with a preset threshold, and determining the trade background authenticity level based on the comparison result.

[0130] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: after obtaining the evaluation score corresponding to each dimension, if the evaluation score is less than a set value, generating evaluation description text and improvement suggestion text, and generating a risk assessment report based on the evaluation description text and improvement suggestion text.

[0131] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:

[0132] Obtain enterprise trade data;

[0133] Identify entities, entity types, and semantic relationships between entities in enterprise trade data; extract trade events from enterprise trade data; and construct a trade knowledge graph based on entities, entity types, semantic relationships, and trade events.

[0134] The authenticity of trade background is quantitatively assessed from multiple dimensions based on a trade knowledge graph, and an assessment score is obtained for each dimension; the level of authenticity of trade background is determined based on the assessment score.

[0135] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: collecting raw trade data from multiple data sources; preprocessing the raw trade data to obtain enterprise trade data; the data preprocessing includes: segmenting, tagging, and naming unstructured text data in the raw trade data; standardizing the format and removing outliers from the structured data in the raw trade data; and associating the same entity from different data sources.

[0136] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: identifying entities in enterprise trade data and the entity types to which the entities belong based on a pre-trained entity recognition model; extracting semantic relationships between entities using dependency parsing and semantic role labeling techniques; and identifying trade events from enterprise trade data using an event extraction model.

[0137] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: assessing the authenticity of entities in the trade knowledge graph by comparing multi-source data to obtain an entity authenticity assessment score; determining the reasonableness of semantic relationships in the trade knowledge graph based on historical data and industry rules to obtain a relationship reasonableness assessment score; verifying the consistency of trade events in the knowledge graph in terms of time sequence and logical order to obtain an event consistency assessment score; and assessing the compliance of trade events in the knowledge graph to obtain a compliance assessment score.

[0138] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: obtaining predefined weights corresponding to each dimension; determining a comprehensive authenticity score based on the predefined weights, entity authenticity assessment score, relationship rationality assessment score, event consistency assessment score, and compliance assessment score; comparing the comprehensive authenticity score with a preset threshold, and determining the trade background authenticity level based on the comparison result.

[0139] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: after obtaining the evaluation score corresponding to each dimension, if the evaluation score is less than a set value, generating evaluation description text and improvement suggestion text, and generating a risk assessment report based on the evaluation description text and improvement suggestion text.

[0140] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0141] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0142] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0143] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A trade background authenticity confirmation method characterized by comprising: The method comprises: acquiring enterprise trade data; identifying entities in the enterprise trade data, entity types to which the entities belong, and semantic relationships between the entities; extracting trade events from the enterprise trade data; and constructing a trade knowledge graph based on the entities, the entity types, the semantic relationships, and the trade events; based on the trade knowledge graph, quantitatively evaluating trade background authenticity from multiple dimensions to obtain evaluation scores of each dimension; and determining a trade background authenticity level based on the evaluation scores.

2. The method of claim 1, wherein, The acquisition of the enterprise trade data comprises: collecting original trade data from multiple data sources; performing data preprocessing on the original trade data to obtain the enterprise trade data; the data preprocessing comprises: performing word segmentation, part-of-speech tagging, and entity naming on unstructured text data in the original trade data, performing format standardization and outlier exclusion on structured data in the original trade data, and associating the same entities from different data sources.

3. The method of claim 1, wherein, identifying entities in the enterprise trade data, entity types to which the entities belong, and semantic relationships between the entities; extracting trade events from the enterprise trade data comprises: identifying entities in the enterprise trade data and entity types to which the entities belong based on a pre-trained entity recognition model; extracting semantic relationships between the entities by using dependency syntax analysis and semantic role labeling technology; identifying trade events from the enterprise trade data by using an event extraction model.

4. The method of claim 1, wherein, The multiple dimensions comprise an entity authenticity dimension, a relationship reasonableness dimension, an event consistency dimension, and a behavior compliance dimension; based on the trade knowledge graph, quantitatively evaluating trade background authenticity from multiple dimensions comprises: evaluating the authenticity of entities in the trade knowledge graph by comparing multi-source data to obtain an entity authenticity evaluation score; determining the reasonableness of semantic relationships in the trade knowledge graph based on historical data and industry rules to obtain a relationship reasonableness evaluation score; verifying the consistency of trade events in the knowledge graph in terms of time sequence and logical order to obtain an event consistency evaluation score; evaluating the compliance of trade events in the knowledge graph to obtain a compliance evaluation score.

5. The method of claim 4, wherein, The determination of the trade background authenticity level based on the evaluation scores comprises: obtaining predefined weights corresponding to each dimension; determining a comprehensive authenticity score based on the predefined weights, the entity authenticity evaluation score, the relationship reasonableness evaluation score, the event consistency evaluation score, and the compliance evaluation score; comparing the comprehensive authenticity score with a preset threshold, and determining a trade background authenticity level according to a comparison result.

6. The method of claim 1, wherein, The method further comprises: after obtaining the evaluation score corresponding to each dimension, generating an evaluation explanation text and an improvement suggestion text in a case where the evaluation score is less than a set value, and generating a risk evaluation report based on the evaluation explanation text and the improvement suggestion text.

7. A trade background authenticity confirmation device characterized by comprising: The device comprises: an acquisition module configured to acquire enterprise trade data; An identification module is configured to identify entities in the enterprise trade data, entity types to which the entities belong, and semantic relationships between the entities; extract trade events from the enterprise trade data; and construct a trade knowledge graph based on the entities, the entity types, the semantic relationships, and the trade events. An evaluation module is configured to quantitatively evaluate trade background authenticity from multiple dimensions based on the trade knowledge graph, to obtain evaluation scores of each dimension, and to determine a trade background authenticity level based on the evaluation scores.

8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that, The processor executes the computer program to implement the steps of the method of any one of claims 1 to 6.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method of any one of claims 1 to 6.

10. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method of any one of claims 1 to 6.

Images