Investment risk identification method and device, equipment, storage medium and program product

Abstract

Embodiments of the present application disclose an investment risk identification method, device, equipment, storage medium and program product, to solve the problem of one-sided and inaccurate investment risk identification results in the prior art. The method comprises: constructing an investment knowledge graph, the investment knowledge graph comprising a plurality of investment entities related to the whole investment process and relationships between different investment entities in the plurality of investment entities; determining a low-dimensional topological representation of the plurality of investment entities according to the investment knowledge graph through topological representation learning; calculating the abnormality degree of the plurality of investment entities according to the low-dimensional topological representation of the plurality of investment entities; and identifying investment risks according to the abnormality degree.

Description

Technical Field

[0001] This application relates to the field of risk identification technology, and in particular to an investment risk identification method, apparatus, equipment, storage medium, and program product. Background Technology

[0002] With the rapid development of the communications industry, communications companies are facing an increasingly complex investment decision-making environment. Specifically, the continuous emergence of new technologies such as 5G, IoT, intelligent computing, and AI has made investment directions more diversified. At the same time, the accelerated pace of technological change, stricter policy supervision, and increased uncertainty in market demand have significantly increased investment risks.

[0003] In existing technologies, investment risks are typically identified using qualitative analysis methods based on expert experience and big data statistical analysis / machine learning methods. However, when using qualitative analysis methods based on expert experience to identify investment risks, the results are prone to subjectivity and lack of timeliness because they rely heavily on expert experience and qualitative analysis. Secondly, when using big data statistical analysis / machine learning methods to identify investment risks, the results are easily biased and inaccurate because they typically only consider single-dimensional data such as financial indicators and project progress.

[0004] Therefore, how to comprehensively and accurately identify investment risks is a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] This application provides an investment risk identification method to address the problem of one-sided and inaccurate investment risk identification results in the prior art.

[0006] This application also provides an investment risk identification device, an electronic device, a computer-readable storage medium, and a computer program product.

[0007] The embodiments of this application adopt the following technical solutions: In a first aspect, embodiments of this application provide a method for identifying investment risks, including: Construct an investment knowledge graph, which includes multiple investment entities related to the entire investment process and the relationships between different investment entities; Based on the investment knowledge graph, low-dimensional topological representations of multiple investment entities are determined through topological representation learning; The degree of anomaly of multiple investment entities is calculated based on their low-dimensional topological representations. Identify investment risks based on the degree of anomaly.

[0008] Optionally, construct an investment knowledge graph, including: Acquire investment data related to the entire investment process, including internal investment data and external environmental data; Entity extraction is performed based on investment data to obtain multiple investment entities; Determine the relationships between different investment entities among multiple investment entities, including at least investment relationships, cooperative relationships, competitive relationships, and / or related constraint relationships; An investment knowledge graph is constructed based on multiple investment entities and their relationships.

[0009] Optionally, low-dimensional topological representations of multiple investment entities are determined through topological representation learning based on the investment knowledge graph, including: The investment knowledge graph is represented as an adjacency matrix, and the matrix element values ​​are used to characterize the strength of the association between different investment entities. A joint optimization objective function is constructed based on the adjacency matrix to solve the target topology matrix, which is used to characterize the low-dimensional topological representation of multiple investment entities. The objective topology matrix is ​​determined by jointly optimizing the objective function, and a low-dimensional topological representation of multiple investment entities is obtained based on the objective topology matrix. The joint optimization objective function includes at least a matrix approximation term, a sparsity constraint term, and a local smoothness constraint term. The matrix approximation term is used to constrain the structural consistency between the target topological matrix and the adjacency matrix, the sparsity constraint term is used to constrain the sparsity of the target topological matrix, and the local smoothness constraint term is used to constrain the robustness of the target topological matrix.

[0010] Optionally, the degree of anomaly of multiple investment entities can be calculated based on their low-dimensional topological representations, including: Based on the low-dimensional topological representation of multiple investment entities, the similarity anomalies and association anomalies of multiple investment entities are calculated respectively; Based on similarity anomalies and association anomalies, the degree of anomaly of multiple investment entities is obtained through an adaptive weighting method.

[0011] Optionally, based on the low-dimensional topological representation of multiple investment entities, similarity anomalies and association anomalies of multiple investment entities are calculated respectively, including: Calculate the topological similarity of multiple investment entities based on their low-dimensional topological representations. Similarity anomalies are determined based on the topological similarity of multiple investment entities and the adaptive anomaly threshold of each investment entity among the multiple investment entities; Based on the correlation strength, topological similarity, and adaptive anomaly threshold of different investment entities in the investment knowledge graph, correlation anomalies are identified.

[0012] Optionally, the method further includes: Before determining similarity anomalies based on the topological similarity of multiple investment entities and the adaptive anomaly threshold of each investment entity among the multiple investment entities, the anomaly threshold of the investment entity is initialized. The anomaly threshold of an investment entity is iteratively updated based on the topological similarity between the investment entity and other investment entities among multiple investment entities. When the preset number of iterations or preset convergence conditions are met, the updated anomaly threshold is determined as the adaptive anomaly threshold for the investment entity.

[0013] Secondly, embodiments of this application provide an investment risk identification device, including a construction module, a determination module, a calculation module, and an identification module, wherein: The construction module is used to build an investment knowledge graph, which includes multiple investment entities related to the entire investment process and the relationships between different investment entities. The determination module is used to determine the low-dimensional topological representation of multiple investment entities based on the investment knowledge graph through topological representation learning; The calculation module is used to calculate the degree of anomaly of multiple investment entities based on their low-dimensional topological representations. The identification module is used to identify investment risks based on the degree of anomaly.

[0014] Thirdly, embodiments of this application provide an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the investment risk identification method described above.

[0015] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the investment risk identification method described above.

[0016] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the investment risk identification method described above.

[0017] The above-described technical solutions adopted in the embodiments of this application can achieve the following beneficial effects: The method provided in this application constructs an investment knowledge graph encompassing multiple investment entities and their interrelationships throughout the entire investment process. Based on this knowledge graph, a low-dimensional topological representation of each investment entity is determined through topological representation learning. Then, the degree of anomaly of each investment entity is calculated based on this low-dimensional topological representation. This approach enables the comprehensive utilization of inter-entity information during risk identification and quantifies investment behavior characteristics at the overall topological structure level. This allows for comprehensive and accurate identification of investment risks, addressing the technical problems of existing technologies that rely on expert experience leading to high subjectivity and poor timeliness, and that analysis based solely on a single dimension results in biased and inaccurate identification. Attached Figure Description

[0018] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments of this application and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 A schematic diagram illustrating the implementation process of an investment risk identification method provided in this application embodiment; Figure 2 This application provides a schematic diagram illustrating the implementation process of a method for determining a low-dimensional topological representation of multiple investment entities in an embodiment of the present application. Figure 3 A schematic diagram illustrating the implementation process of a method for determining similarity anomalies and association anomalies of investment entities, provided in an embodiment of this application; Figure 4 This application provides a schematic diagram of the specific structure of an investment risk identification device. Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0020] As will be known to those skilled in the art, with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0021] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such terms are interchangeable where appropriate; this is merely a way of distinguishing objects with the same attributes in the embodiments of this application. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion, so that a process, method, system, product, or apparatus that comprises a series of elements is not necessarily limited to those elements but may include other elements not explicitly listed or inherent to those processes, methods, products, or apparatuses.

[0022] It should be understood that the training and prediction processes of the AI ​​models involved in the various embodiments of this specification all adhere to multiple legal and compliant principles, including legal data sources, compliant data content, compliant data governance, compliant training objectives and schemes, compliant training processes, compliant training environments and tools, and compliant ethical verification of training results, and comply with the requirements of Article 5 of the Patent Law. Among them: Data source legitimacy: All datasets used for AI model training were obtained through legal means, covering three categories: publicly authorized data, data authorized by partners, and self-collected compliant data. Publicly authorized data comes from compliant data sources following open-source licenses such as Apache 2.0, with complete copyright attribution and authorization scope clearly marked, and no unauthorized open-source code or data reuse. Data authorized by partners has been subject to formal data usage agreements, clearly defining the scope, duration, and confidentiality obligations, and possessing a complete authorization chain. For self-collected data involving personal information, strict informed consent procedures have been followed, and anonymization processes (including but not limited to field masking, feature anonymization, and differential privacy technology applications) have been implemented to remove personally identifiable information, fully complying with the requirements of relevant laws and regulations such as the "Interim Measures for the Administration of Generative Artificial Intelligence Services" and the "Personal Information Protection Law."

[0023] Data content compliance: The AI ​​model's dataset undergoes multiple screenings and cleaning processes to remove all content that may violate social morality or harm public interests. It contains no obscene, pornographic, violent, discriminatory, or information that endangers national or public safety, nor does it involve the illegal acquisition or use of genetic resources. For data in sensitive fields (such as healthcare and finance), an additional privacy-preserving computation module (including federated learning and secure multi-party computation technologies) ensures that the data is "usable but not visible," avoiding compliance risks during the original data transmission process and ensuring that the data application scenarios and uses comply with public order and good morals and industry regulatory requirements.

[0024] Data governance norms: A complete data traceability system is established during the AI ​​model training process to automatically record the source, collection time, annotation process, cleaning rules, and permission allocation of training data, generating traceable compliance reports to ensure that the data is verifiable throughout its entire lifecycle. The dataset annotation process for AI models is completed by a professional human R&D team, clearly defining the proportion of human creative contributions and avoiding reliance on AI-generated data that has not undergone substantial human modification, thus meeting the examination requirements for "human main contributions" in AI patent applications.

[0025] Training objectives and plans are compliant: The AI ​​model training objective focuses on risk identification and early warning throughout the entire investment process in the communications industry. Based on investment knowledge graphs and topological representations of investment entities, it detects anomalies in investment behavior to assist investment management departments in timely identifying potential risks, improving decision-making efficiency and risk control capabilities. The training scheme and final output results do not violate any mandatory provisions of laws and administrative regulations, do not harm the public interest or the legitimate rights and interests of others, and do not pose any potential risks of being used for illegal activities, infringing on privacy, or disrupting public safety. It strictly adheres to the ethical principle of "intelligent for good".

[0026] Training process compliance: A closed-loop training framework is adopted to ensure compliance and controllability of the training process. The specific process is as follows: First, training samples are obtained through compliant data sources. After the aforementioned data cleaning and desensitization, they are input into the neural network model to generate preliminary training results. Second, an expert system is introduced to verify the preliminary results. Based on preset rules and human expert experience, the feasibility of the results is evaluated, and outputs that may pose ethical risks or compliance hazards are corrected (such as removing decision-making logic that violates public order and good morals, and adjusting model parameters that do not comply with safety regulations). Finally, the loss function weights are dynamically optimized based on expert system feedback to strengthen the model's learning of compliant results, avoid overfitting errors or non-compliant labels, and form a closed-loop control of "data input - model training - expert verification - parameter optimization - result feedback" to ensure that the entire training process complies with A5 ethical review requirements.

[0027] Training environment and tool compliance: AI model training is implemented using nationally licensed chips and a compliant training platform. All open-source frameworks and components used in the training process have obtained their corresponding licenses, and copyright statements and patent citation information are fully retained, with no instances of infringement or reuse. The training environment is built using virtual devices (containers / virtual machines) with fixed random seeds and initial parameter configurations to ensure the reproducibility of the training process. Furthermore, through access control and operation log recording, risks such as data leakage and parameter tampering during training are prevented, ensuring the security and compliance of the training process.

[0028] Training results ethical verification compliance: After the model is trained, it undergoes additional third-party ethical compliance assessment and algorithm filing review to verify that the model output does not violate social morality or harm public interests. For potentially sensitive scenarios (such as public services and intelligent decision-making), a special result verification mechanism is established to ensure that the model always complies with Article 5 of the Patent Law and relevant laws and regulations in practical applications.

[0029] In summary, the data and training process used in the AI ​​model of this specification strictly comply with the relevant provisions of Article 5 of the Patent Law and the Patent Examination Guidelines (2023 Edition), and there are no violations of laws, social ethics, public interests, or illegal use of genetic resources. It fully meets the compliance requirements for patent authorization.

[0030] Example 1 To address the problem of one-sided and inaccurate investment risk identification results in existing technologies, this application provides an investment risk identification method.

[0031] The execution subject of this method can be various types of computing devices, or it can be an application or app installed on the computing device. The computing device can be a user terminal such as a mobile phone, tablet computer, or smart wearable device, or it can be a server.

[0032] For ease of description, this application uses a server as the execution subject of the method in its embodiments to illustrate the method. Those skilled in the art will understand that this embodiment uses a server as an example to describe the method, which is merely an illustrative example and does not limit the scope of protection of the corresponding claims.

[0033] Specifically, the implementation flow of the method provided in this application embodiment is as follows: Figure 1 As shown, it includes the following steps: Step 102: Construct an investment knowledge graph. The investment knowledge graph includes multiple investment entities related to the entire investment process and the relationships between different investment entities.

[0034] In this embodiment, when constructing an investment knowledge graph, investment data related to the entire investment process can be acquired first. This entire investment process may include stages such as project initiation, feasibility studies, budgeting, procurement, contracts, payments, acceptance, and operation. The investment data includes internal investment data and external environmental data. Then, entity extraction is performed based on the investment data to obtain multiple investment entities. Next, the relationships between different investment entities are determined, including at least investment relationships, cooperative relationships, competitive relationships, and / or relationship constraints. Finally, an investment knowledge graph is constructed based on the multiple investment entities and their relationships.

[0035] Internal investment data includes at least structured data such as corporate internal investment records, financial reports, and project management documents; external environment data includes at least unstructured data such as industry policy documents, market research reports, and news media information.

[0036] In some embodiments, investment-related data can first be collected from multiple heterogeneous data sources. This investment data includes at least structured data such as internal corporate investment records, financial reports, and project management documents, as well as unstructured data such as industry policy documents, market research reports, and news media information.

[0037] Secondly, information extraction technology is used to identify investment-related entities from investment data. These entities include, but are not limited to, investment entities (such as telecommunications companies and investment institutions), investment projects (such as 5G base station construction and data center construction), investment technologies (such as 5G technology and cloud computing technology), policies and regulations (such as spectrum allocation policies and industry access regulations), and market environments (such as regional market demand and competitive landscape).

[0038] Then, based on semantic analysis, the relationships between different investment entities are determined. These relationships include explicit relationships (such as direct investment relationships and equity relationships) and implicit relationships (such as technology dependence relationships and policy influence relationships). Specifically, they can include investment relationships, cooperative relationships, competitive relationships, and related constraint relationships.

[0039] Among them, the relationship of association constraint is used to describe the restrictions that an investment entity is subject to under certain conditions, rules or boundaries, or the constraint relationship with certain compliance / policy / qualification / funding / process requirements.

[0040] Finally, the identified investment entities and relationships are organized to form a structured investment knowledge graph, which is then stored and managed using graph databases such as Neo4j.

[0041] Optionally, to ensure that heterogeneous data can be expressed in a related manner within the same semantic space, knowledge graph technologies such as Resource Description Framework (RDF) and Web Ontology Language (OWL) can be used to semantically represent and associate investment data. The constructed investment knowledge graph can then be stored in a graph database for semantic retrieval and reasoning using query languages ​​such as Cypher, thereby uncovering the implicit relationships between investment entities and providing comprehensive knowledge support for subsequent risk identification.

[0042] Step 104: Determine the low-dimensional topological representations of multiple investment entities through topological representation learning based on the investment knowledge graph.

[0043] In the embodiments of this application, such as Figure 2As shown, the low-dimensional topological representation of multiple investment entities can be determined through the following steps 202-206: Step 202: Represent the investment knowledge graph as an adjacency matrix. The matrix element values ​​of the adjacency matrix are used to characterize the correlation strength between different investment entities.

[0044] In some embodiments, the adjacency matrix of the investment knowledge graph can be determined based on the association strength between different investment entities among multiple investment entities. Specifically, assuming there are N investment entities (including investment entities, projects, technologies, policies, etc.), an N×N adjacency matrix A can be constructed, where the matrix element Aij in adjacency matrix A represents the association strength between the i-th investment entity and the j-th investment entity.

[0045] Optionally, the calculation of correlation strength can comprehensively consider factors such as relationship type, relationship attributes (e.g., investment amount, policy influence), and relationship timeliness (recent relationships are given higher weight). For example, for two 5G base station construction projects, if they are affected by the same policy and implemented in the same area, the correlation strength between these two 5G base station construction projects can be considered strong. Therefore, a larger correlation strength can be set, such as 0.75.

[0046] It should be noted that the methods for determining the association strength listed above are merely illustrative examples in the embodiments of this application and do not impose any limitations on the embodiments of this application.

[0047] Step 204: Construct a joint optimization objective function based on the adjacency matrix to solve the target topology matrix. The target topology matrix is ​​used to characterize the low-dimensional topological representation of multiple investment entities. The joint optimization objective function includes at least a matrix approximation term, a sparsity constraint term, and a local smoothness constraint term.

[0048] The matrix approximation term is used to constrain the structural consistency between the target topology matrix and the adjacency matrix. In other words, it ensures that the solved target topology matrix retains as much of the overall topological structure of the investment knowledge graph as possible.

[0049] Sparsity constraint terms are used to constrain the sparsity of low-dimensional topological representations.

[0050] Local smoothness constraint terms are used to constrain the robustness of low-dimensional topological representations.

[0051] In this embodiment, the target dimension k of the low-dimensional topological representation can be determined first, and an N×k initial topological matrix T can be initialized based on the target dimension k, where each row Ti of T represents the k-dimensional low-dimensional topological representation vector of the i-th investment entity. Then, a joint optimization objective function for solving the target topological matrix is ​​constructed based on the adjacency matrix and the initial topological matrix, as follows:

[0052] in, Represents the adjacency matrix; Represents the initial topology matrix; Representing the adjacency matrix The matrix elements represent the correlation strength between the i-th investment entity and the j-th investment entity. This represents the topological representation vector of the i-th row of the initial topological matrix, which is also the topological representation vector of the i-th investment entity. This represents the topological representation vector of the j-th row of the initial topological matrix, which is also the j-th investment entity. Describing the Frobenius norm, Describing the L1 norm, This represents the L2 norm. and This is a preset balance factor. Represents the matrix approximation terms; Represents the sparsity constraint term; This represents the local smoothness constraint term.

[0053] Step 206: Determine the target topology matrix by jointly optimizing the objective function, and obtain the low-dimensional topological representation of multiple investment entities based on the target topology matrix.

[0054] After obtaining the above joint optimization objective function, the joint optimization objective function can be solved by numerical optimization algorithms such as gradient descent and L-BFGS, and the initial topology matrix T can be iteratively updated until it converges to obtain the target topology matrix T*. Each row of the target topology matrix T* is the low-dimensional topology representation vector of the corresponding investment entity.

[0055] It should be noted that the gradient descent and L-BFGS optimization algorithms listed above are merely illustrative examples in the embodiments of this application and do not limit the embodiments of this application.

[0056] In this embodiment, a low-dimensional topological representation of investment entities is determined using a topological representation learning method. On the one hand, this method maps the various types of relationships between investment entities in the investment knowledge graph and their global topological structure information to a low-dimensional vector space, thereby achieving a compact expression of the structural features of investment entities. On the other hand, by introducing matrix approximation terms, sparsity constraint terms, and local smoothness constraint terms into the joint optimization objective function, the interference of noisy relationships and random behaviors on the representation results can be effectively suppressed, improving the robustness and generalization ability of the low-dimensional topological representation.

[0057] Furthermore, low-dimensional topological representation transforms the original high-dimensional and complex graph structure into a fixed-dimensional vector representation, which can significantly reduce the computational complexity of subsequent anomaly detection and risk identification, improve real-time processing efficiency, and thus be more suitable for investment risk identification and early warning in the context of continuous data updates throughout the investment process.

[0058] Step 106: Calculate the degree of anomaly of multiple investment entities based on their low-dimensional topological representations.

[0059] In this embodiment of the application, when determining the degree of anomaly, the similarity anomaly and the association anomaly of the multiple investment entities can be calculated first based on the low-dimensional topological representation of the multiple investment entities; then, based on the similarity anomaly and the association anomaly, the degree of anomaly of the multiple investment entities can be obtained by an adaptive weighting method.

[0060] Among them, similarity anomalies are used to characterize the degree of similarity between the topological structure of an investment entity and its surrounding entities in its low-dimensional topological space. The lower the similarity, the higher the degree of anomaly. Association anomalies are used to characterize the degree of deviation between the actual association of an investment entity in the investment knowledge graph and its expected association in the low-dimensional topological space. The greater the deviation, the higher the degree of anomaly.

[0061] In some embodiments, such as Figure 3 As shown, similarity anomalies and association anomalies of investment entities can be identified by following these steps: Step 302: Calculate the topological similarity of multiple investment entities based on their low-dimensional topological representations.

[0062] Topological similarity measures the proximity of different investment entities in the topological space. Optionally, topological similarity can be calculated using methods such as cosine similarity or Euclidean distance. For example:

[0063] in, This represents the topological similarity between investment entity i and investment entity j. Represents the i-th row vector of the target topology matrix T*; Let represent the j-th row vector of the target topology matrix T*.

[0064] Step 304: Based on the topological similarity of multiple investment entities and the adaptive anomaly threshold of each investment entity among the multiple investment entities, determine the similarity anomaly.

[0065] Among them, the adaptive anomaly threshold is a dynamically adjusted threshold parameter used to distinguish between the normal and abnormal states of investment entities. It can be automatically adjusted according to the historical false alarm rate to maintain the preset target false alarm rate level.

[0066] In this embodiment of the application, similarity anomalies can be obtained according to the following calculation method:

[0067] in, This indicates an anomaly in the similarity of investment entity i; Weight parameters representing topological similarity; Represents the adaptive anomaly threshold for investment entity i; This represents the total number of investment entities involved in the calculation; This indicates an indicator function that takes the value 1 if the condition within the parentheses is true, and 0 otherwise.

[0068] In some implementations, when determining the adaptive anomaly threshold of an investment entity, the anomaly threshold of the investment entity can be initialized first. Then, based on the topological similarity between the investment entity and other investment entities among multiple investment entities, the anomaly threshold of the investment entity is iteratively updated. When a preset number of iterations or a preset convergence condition is met, the iteratively updated anomaly threshold is determined as the adaptive anomaly threshold of the investment entity. The calculation expression for the iterative update is as follows:

[0069] in, This represents the adaptive anomaly threshold of investment entity i at the (t+1)th iteration update; This represents the adaptive anomaly threshold for investment entity i at the t-th iteration update; This represents the preset learning rate, which is used to control the adjustment range for each iteration update.

[0070] Step 306: Determine association anomalies based on the association strength, topological similarity, and adaptive anomaly threshold of each investment entity in the investment knowledge graph.

[0071] In this embodiment of the application, the correlation anomalies of each investment entity can be calculated using the following formula:

[0072] in, This indicates an abnormal association between investment entity i and other entities. The weight parameter represents the correlation anomaly.

[0073] In summary, in one implementation, the degree of abnormality of investment entity i can be calculated according to the following formula:

[0074] Indicates the degree of abnormality of investment entity i; This indicates an indicator function that takes the value 1 if the condition within the parentheses is true, and 0 otherwise.

[0075] After calculating the degree of abnormality of each investment entity in the above manner, the degree of abnormality of each investment entity can be weighted and summed to obtain the degree of abnormality of multiple investment entities.

[0076] Step 108: Identify investment risks based on the degree of anomaly.

[0077] In some embodiments, the degree of anomaly can be compared with preset risk assessment criteria; these risk assessment criteria may include an investment risk anomaly threshold, risk level classification rules, or a combination thereof. When the degree of anomaly reaches or exceeds the investment risk anomaly threshold, the investment entity is determined to have investment risk, and a risk identification result corresponding to the investment entity is output. The risk identification result may include a risk identifier, risk level, and / or warning information.

[0078] Alternatively, when the degree of anomaly is below the investment risk anomaly threshold, the investment entity can be identified as not having triggered investment risk or in a low-risk state. In practical applications, the investment risk anomaly threshold can be a fixed threshold, or it can be a dynamic threshold determined based on the historical anomaly distribution and a self-learning update strategy, so that the risk assessment boundary can be dynamically adjusted with changes in investment patterns, thereby improving the accuracy and adaptability of investment risk identification.

[0079] In addition, after outputting the risk identification results, the investment entities that triggered the risk can be traced back to their origins based on the investment knowledge graph. By retrieving related entities that are strongly associated with the investment entity and have low topological similarity, the association link information used to explain the cause of the risk can be generated, thereby improving the interpretability and verifiability of the risk identification results.

[0080] In some embodiments, in order to enable timely handling after identifying abnormal investment behavior and improve the interpretability of risk identification results, an early warning can be automatically triggered when a potential abnormal investment event is identified. An investment knowledge graph can be used to trace the abnormal behavior, and key influencing factors of the abnormality can be located through graph traversal and pattern matching operations of the graph database and a risk explanation report can be generated. In this way, rapid alerts and cause tracing of abnormal investment events can be achieved, which makes it easier for business personnel to take timely countermeasures and verify the causes of risks, thereby improving the timeliness and reliability of investment risk management.

[0081] Optionally, to enhance the visualization and business usability of investment risk identification results, some embodiments may develop a web front-end system and design multi-dimensional risk indicator dashboards to intuitively present information such as the degree of anomaly of investment entities and risk transmission paths. Simultaneously, visualization libraries such as D3.js and Echarts can be used to enable interactive exploration of investment knowledge graphs, as well as analytical functions such as investment scenario simulation and predictive warnings. The back-end uses the Spring Boot framework to provide RESTful APIs for data interaction with the front-end. This allows for intuitive display and interactive analysis of risk information, facilitating business personnel to quickly locate risk sources, understand risk transmission mechanisms, and support decision-making, thereby improving the efficiency and accuracy of investment risk management.

[0082] The method provided in this application constructs an investment knowledge graph encompassing multiple investment entities and their interrelationships throughout the entire investment process. Based on this knowledge graph, a low-dimensional topological representation of each investment entity is determined through topological representation learning. Then, the degree of anomaly of each investment entity is calculated based on this low-dimensional topological representation. This approach enables the comprehensive utilization of inter-entity information during risk identification and quantifies investment behavior characteristics at the overall topological structure level. This allows for comprehensive and accurate identification of investment risks, addressing the technical problems of existing technologies that rely on expert experience leading to high subjectivity and poor timeliness, and that analysis based solely on a single dimension results in biased and inaccurate identification.

[0083] Example 2 To address the problem of incomplete and inaccurate investment risk identification results in existing technologies, this application provides an investment risk identification device, the specific structure of which is shown in the schematic diagram below. Figure 4 As shown, it includes a construction module 41, a determination module 42, a calculation module 43, and a recognition module 44. The functions of each module are as follows: Module 41 is used to construct an investment knowledge graph, which includes multiple investment entities related to the entire investment process and the relationships between different investment entities among the multiple investment entities; The determination module 42 is used to determine the low-dimensional topological representation of the multiple investment entities based on the investment knowledge graph through topological representation learning; Calculation module 43 is used to calculate the degree of anomaly of the plurality of investment entities based on the low-dimensional topological representation of the plurality of investment entities; The identification module 44 is used to identify investment risks based on the degree of anomaly.

[0084] Optional, building module 41, for: Acquire investment data related to the entire investment process, including internal investment data and external environmental data; Entity extraction is performed based on investment data to obtain multiple investment entities; Determine the relationships between different investment entities among multiple investment entities, including at least investment relationships, cooperative relationships, competitive relationships, and / or related constraint relationships; An investment knowledge graph is constructed based on multiple investment entities and their relationships.

[0085] Optionally, module 42 is used for: The investment knowledge graph is represented as an adjacency matrix, and the matrix element values ​​are used to characterize the strength of the association between different investment entities. A joint optimization objective function is constructed based on the adjacency matrix to solve the target topology matrix, which is used to characterize the low-dimensional topological representation of multiple investment entities. The objective topology matrix is ​​determined by jointly optimizing the objective function, and a low-dimensional topological representation of multiple investment entities is obtained based on the objective topology matrix. The joint optimization objective function includes at least a matrix approximation term, a sparsity constraint term, and a local smoothness constraint term. The matrix approximation term is used to constrain the structural consistency between the target topological matrix and the adjacency matrix, the sparsity constraint term is used to constrain the sparsity of the target topological matrix, and the local smoothness constraint term is used to constrain the robustness of the target topological matrix.

[0086] Optionally, the calculation module 43 includes: The first computing unit is used to calculate the similarity anomalies and association anomalies of multiple investment entities based on their low-dimensional topological representations. The second calculation unit is used to obtain the degree of anomaly of multiple investment entities through an adaptive weighting method based on similarity anomalies and correlation anomalies.

[0087] Optionally, the first computing unit is used for: Calculate the topological similarity of multiple investment entities based on their low-dimensional topological representations. Similarity anomalies are determined based on the topological similarity of multiple investment entities and the adaptive anomaly threshold of each investment entity among the multiple investment entities; Based on the correlation strength, topological similarity, and adaptive anomaly threshold of different investment entities in the investment knowledge graph, correlation anomalies are identified.

[0088] Optionally, the investment risk identification device further includes an adaptive anomaly threshold determination module, used for: Initialize the abnormal threshold for investment entities; The anomaly threshold of an investment entity is iteratively updated based on the topological similarity between the investment entity and other investment entities among multiple investment entities. When the preset number of iterations or preset convergence conditions are met, the updated anomaly threshold is determined as the adaptive anomaly threshold for the investment entity.

[0089] The apparatus provided in this application constructs an investment knowledge graph encompassing multiple investment entities and their interrelationships throughout the entire investment process. Based on this knowledge graph, it determines low-dimensional topological representations of multiple investment entities through topological representation learning. Then, it calculates the degree of anomaly of multiple investment entities based on these low-dimensional topological representations. This allows for the comprehensive utilization of the correlation information between investment entities during risk identification and the quantification of investment behavior characteristics at the overall topological structure level. Consequently, it enables comprehensive and accurate identification of investment risks, addressing the technical problems of existing technologies that rely on expert experience, leading to strong subjectivity and poor timeliness, as well as the one-sided and inaccurate identification caused by analysis based on only a single dimension of data.

[0090] Example 3 Figure 5 To illustrate the hardware structure of an electronic device according to various embodiments of this application, the electronic device 500 includes, but is not limited to, components such as: a radio frequency unit 501, a network module 502, an audio output unit 503, an input unit 504, a sensor 505, a display unit 506, a user input unit 507, an interface unit 508, a memory 509, a processor 510, and a power supply 511. Those skilled in the art will understand that... Figure 5 The electronic device structures shown are not intended to limit the electronic device. An electronic device may include more or fewer components than shown, or combine certain components, or have different component arrangements. In the embodiments of this application, the electronic device includes, but is not limited to, mobile phones, tablets, laptops, PDAs, in-vehicle terminals, wearable devices, and pedometers.

[0091] The processor 510 is used to construct an investment knowledge graph, which includes multiple investment entities related to the entire investment process and the relationships between different investment entities; it determines the low-dimensional topological representation of multiple investment entities through topological representation learning based on the investment knowledge graph; it calculates the degree of anomaly of multiple investment entities based on the low-dimensional topological representation of multiple investment entities; and it identifies investment risks based on the degree of anomaly.

[0092] The memory 509 is used to store a computer program that can run on the processor 510, which, when executed by the processor 510, performs the functions described above by the processor 510.

[0093] It should be understood that, in this embodiment, the radio frequency unit 501 can be used for receiving and transmitting signals during information transmission or calls. Specifically, it receives downlink data from the base station and processes it with the processor 510; additionally, it transmits uplink data to the base station. Typically, the radio frequency unit 501 includes, but is not limited to, an antenna, at least one amplifier, a transceiver, a coupler, a low-noise amplifier, a duplexer, etc. Furthermore, the radio frequency unit 501 can also communicate with networks and other devices through a wireless communication system.

[0094] The electronic device provides users with wireless broadband internet access through the network module 502, such as helping users send and receive emails, browse web pages, and access streaming media.

[0095] The audio output unit 503 can convert audio data received by the radio frequency unit 501 or the network module 502 or stored in the memory 509 into audio signals and output them as sound. Furthermore, the audio output unit 503 can also provide audio output related to specific functions performed by the electronic device 500 (e.g., call signal reception sound, message reception sound, etc.). The audio output unit 503 includes a speaker, a buzzer, and a receiver, etc.

[0096] Input unit 504 is used to receive audio or video signals. Input unit 504 may include a graphics processing unit (GPU) 5041 and a microphone 5042. GPU 5041 processes image data of still images or videos acquired by an image capture device (such as a camera) in video capture mode or image capture mode. The processed image frames can be displayed on display unit 506. The image frames processed by GPU 5041 can be stored in memory 509 (or other storage medium) or transmitted via radio frequency unit 501 or network module 502. Microphone 5042 can receive sound and process such sound into audio data. The processed audio data can be converted into a format that can be transmitted to a mobile communication base station via radio frequency unit 501 in telephone call mode.

[0097] The electronic device 500 also includes at least one sensor 505, such as a light sensor, a motion sensor, and other sensors. Specifically, the light sensor includes an ambient light sensor and a proximity sensor. The ambient light sensor can adjust the brightness of the display panel 5061 according to the ambient light level, and the proximity sensor can turn off the display panel 5061 and / or backlight when the electronic device 500 is moved to the ear. As a type of motion sensor, an accelerometer sensor can detect the magnitude of acceleration in various directions (generally three axes). When stationary, it can detect the magnitude and direction of gravity and can be used to identify the posture of the electronic device (such as landscape / portrait switching, related games, magnetometer posture calibration), vibration recognition related functions (such as pedometer, tapping), etc. The sensor 505 may also include a fingerprint sensor, pressure sensor, iris sensor, molecular sensor, gyroscope, barometer, hygrometer, thermometer, infrared sensor, etc., which will not be described in detail here.

[0098] The display unit 506 is used to display information input by the user or information provided to the user. The display unit 506 may include a display panel 5061, which may be configured in the form of a liquid crystal display (LCD), an organic light-emitting diode (OLED), or the like.

[0099] User input unit 507 can be used to receive input numerical or character information, and to generate key signal inputs related to user settings and function control of electronic devices. Specifically, user input unit 507 includes a touch panel 5071 and other input devices 5072. Touch panel 5071, also known as a touch screen, can collect touch operations performed by the user on or near it (such as operations performed by the user using a finger, stylus, or any suitable object or accessory on or near touch panel 5071). Touch panel 5071 may include two parts: a touch detection device and a touch controller. The touch detection device detects the user's touch position and the signal generated by the touch operation, and transmits the signal to the touch controller; the touch controller receives touch information from the touch detection device, converts it into touch point coordinates, and sends it to the processor 510, which receives and executes commands from the processor 510. In addition, touch panel 5071 can be implemented using various types such as resistive, capacitive, infrared, and surface acoustic wave. Besides touch panel 5071, user input unit 507 may also include other input devices 5072. Specifically, other input devices 5072 may include, but are not limited to, physical keyboards, function keys (such as volume control buttons, power buttons, etc.), trackballs, mice, joysticks, etc., which will not be described in detail here.

[0100] Furthermore, the touch panel 5071 can cover the display panel 5061. When the touch panel 5071 detects a touch operation on or near it, it transmits the information to the processor 510 to determine the type of touch event. Subsequently, the processor 510 provides corresponding visual output on the display panel 5061 based on the type of touch event. Although in Figure 5 In this embodiment, the touch panel 5071 and the display panel 5061 are two independent components to realize the input and output functions of the electronic device. However, in some embodiments, the touch panel 5071 and the display panel 5061 can be integrated to realize the input and output functions of the electronic device. The specific implementation is not limited here.

[0101] Interface unit 508 serves as an interface for connecting external devices to electronic device 500. For example, external devices may include a wired or wireless headphone port, an external power supply (or battery charger) port, a wired or wireless data port, a memory card port, a port for connecting a device with an identification module, an audio input / output (I / O) port, a video I / O port, a headphone port, and so on. Interface unit 508 can be used to receive input (e.g., data, power, etc.) from external devices and transmit the received input to one or more components within electronic device 500, or it can be used to transmit data between electronic device 500 and external devices.

[0102] The memory 509 can be used to store software programs and various data. The memory 509 may primarily include a program storage area and a data storage area. The program storage area may store the operating system, applications required for at least one function (such as sound playback, image playback, etc.), etc.; the data storage area may store data created based on the use of the mobile phone (such as audio data, phonebook, etc.). Furthermore, the memory 509 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0103] The processor 510 is the control center of the electronic device. It connects various parts of the electronic device via various interfaces and lines. By running or executing software programs and / or modules stored in the memory 509, and by calling data stored in the memory 509, it performs various functions and processes data, thereby providing overall monitoring of the electronic device. The processor 510 may include one or more processing units; preferably, the processor 510 may integrate an application processor and a modem processor. The application processor mainly handles the operating system, user interface, and applications, while the modem processor mainly handles wireless communication. It is understood that the modem processor may not be integrated into the processor 510.

[0104] The electronic device 500 may also include a power supply 511 (such as a battery) that supplies power to various components. Preferably, the power supply 511 can be logically connected to the processor 510 through a power management system, thereby enabling functions such as managing charging, discharging, and power consumption through the power management system.

[0105] In addition, the electronic device 500 includes some functional modules not shown, which will not be described in detail here.

[0106] Preferably, this application embodiment also provides an electronic device, including a processor 510, a memory 509, and a computer program stored in the memory 509 and executable on the processor 510. When the computer program is executed by the processor 510, it implements the various processes of the above-described investment risk identification method embodiment and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0107] This application also provides a computer-readable storage medium storing a computer program. When executed by a processor, this computer program implements the various processes of the above-described investment risk identification method embodiments and achieves the same technical effects. To avoid repetition, it will not be described again here. The computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0108] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0109] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0110] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0111] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0112] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0113] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0114] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0115] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0116] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for identifying investment risks, characterized in that, include: Construct an investment knowledge graph, which includes multiple investment entities related to the entire investment process and the relationships between different investment entities; Based on the investment knowledge graph, low-dimensional topological representations of the multiple investment entities are determined through topological representation learning; The degree of anomaly of the multiple investment entities is calculated based on the low-dimensional topological representation of the multiple investment entities; Investment risks are identified based on the degree of anomaly.

2. The method as described in claim 1, characterized in that, The construction of the investment knowledge graph includes: Acquire investment data related to the entire investment process, including internal investment data and external environment data; Based on the investment data, entity extraction is performed to obtain the multiple investment entities; Determine the relationships between different investment entities among the plurality of investment entities, wherein the relationships include at least investment relationships, cooperative relationships, competitive relationships and / or relationship constraints; The investment knowledge graph is constructed based on the multiple investment entities and the relationships between them.

3. The method as described in claim 1 or 2, characterized in that, The step of determining the low-dimensional topological representation of the multiple investment entities through topological representation learning based on the investment knowledge graph includes: The investment knowledge graph is represented as an adjacency matrix, and the matrix element values ​​of the adjacency matrix are used to characterize the correlation strength between the different investment entities; Based on the adjacency matrix, a joint optimization objective function is constructed to solve the target topology matrix, wherein the topology matrix is ​​used to characterize the low-dimensional topological representation of the multiple investment entities; The target topology matrix is ​​determined by the joint optimization objective function, and the low-dimensional topological representation of the plurality of investment entities is obtained based on the target topology matrix; The joint optimization objective function includes at least a matrix approximation term, a sparsity constraint term, and a local smoothness constraint term. The matrix approximation term is used to constrain the structural consistency between the target topology matrix and the adjacency matrix, the sparsity constraint term is used to constrain the sparsity of the target topology matrix, and the local smoothness constraint term is used to constrain the robustness of the target topology matrix.

4. The method according to any one of claims 1 to 3, characterized in that, The step of calculating the anomaly degree of the multiple investment entities based on their low-dimensional topological representations includes: Based on the low-dimensional topological representation of the multiple investment entities, the similarity anomalies and association anomalies of the multiple investment entities are calculated respectively; Based on the similarity anomalies and the association anomalies, the degree of anomalies of the multiple investment entities is obtained through an adaptive weighting method.

5. The method as described in claim 4, characterized in that, The calculation of similarity anomalies and association anomalies of the multiple investment entities based on the low-dimensional topological representation of the multiple investment entities includes: Based on the low-dimensional topological representation of the multiple investment entities, the topological similarity of the multiple investment entities is calculated; The similarity anomaly is determined based on the topological similarity of the multiple investment entities and the adaptive anomaly threshold of each investment entity among the multiple investment entities; The association anomalies are determined based on the association strength between different investment entities in the investment knowledge graph, the topological similarity, and the adaptive anomaly threshold of each investment entity.

6. The method as described in claim 5, characterized in that, Before the topological similarity based on the plurality of investment entities and the adaptive anomaly threshold for each investment entity among the plurality of investment entities, the method further includes: Initialize the abnormal threshold of the investment entity; Based on the topological similarity between the investment entity and other investment entities among the plurality of investment entities, the anomaly threshold of the investment entity is iteratively updated. When the preset number of iterations or the preset convergence condition is met, the updated anomaly threshold is determined as the adaptive anomaly threshold of the investment entity.

7. An investment risk identification device, characterized in that, It includes a construction module, a determination module, a calculation module, and a recognition module, among which: A construction module is used to build an investment knowledge graph, which includes multiple investment entities related to the entire investment process and the relationships between different investment entities among these multiple investment entities. The determination module is used to determine the low-dimensional topological representation of the multiple investment entities based on the investment knowledge graph through topological representation learning; A calculation module is used to calculate the degree of anomaly of the plurality of investment entities based on the low-dimensional topological representation of the plurality of investment entities; An identification module is used to identify investment risks based on the degree of anomaly.

8. An electronic device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the investment risk identification method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the investment risk identification method as described in any one of claims 1 to 6.

10. A computer program product, characterized in that, It includes a computer program that, when executed by a processor, implements the investment risk identification method according to any one of claims 1 to 6.

Images