Enterprise asset management strategy generation method and device, electronic equipment and medium

Abstract

The invention discloses an enterprise asset management strategy generation method and apparatus, an electronic device and a medium. The method comprises the steps of obtaining asset flow data of a target enterprise, customer behavior data of the target enterprise and market data of a target industry; analyzing the asset flow data of the target enterprise through an asset flow prediction model, and outputting a first analysis result by the asset flow prediction model; analyzing the market data of the target industry through a market fluctuation prediction model, and outputting a second analysis result by the market fluctuation prediction model; analyzing the customer behavior data of the target enterprise through a customer behavior analysis model, and outputting a third analysis result by the customer behavior analysis model; and generating an asset management strategy of the target enterprise based on the first analysis result, the second analysis result and the third analysis result. The method can be applied to the business fields of financial science and technology, medical health, old-age care and the like, and the rationality of an asset management strategy is improved through multi-dimensional data analysis and model prediction.

Description

Technical Field

[0001] The present application relates to the field of financial technology, and in particular to a method, device, electronic device and medium for generating an enterprise asset management strategy. Background Art

[0002] A company's asset management is a core component of asset-liability management and directly impacts its solvency and market competitiveness. A company's asset allocation strategy, a crucial foundation for asset management, can optimize its asset allocation, thereby enhancing its solvency and market competitiveness. However, related technologies primarily rely on historical data and simple statistical models, supplemented by manual experience, to formulate asset allocation strategies, resulting in poor rationality. Summary of the Invention

[0003] The main purpose of the embodiments of the present application is to propose a method, device, electronic device and medium for generating an enterprise asset management strategy, aiming to solve the problem that the enterprise asset allocation strategy formulated by the existing enterprise asset management strategy generation method is less rational.

[0004] To achieve the above objectives, a first aspect of an embodiment of the present application provides a method for generating an enterprise asset management strategy, the method comprising:

[0005] Obtaining asset flow data of a target enterprise, customer behavior data of the target enterprise, and market data of a target industry, where the target industry is the industry to which the target enterprise belongs;

[0006] Analyzing the asset flow data of the target enterprise using a pre-trained asset flow prediction model, and outputting a first analysis result by the asset flow prediction model, wherein the first analysis result is used to indicate the asset liquidity of the target enterprise;

[0007] Analyzing the market data of the target industry using a pre-trained market fluctuation prediction model, and outputting a second analysis result by the market fluctuation prediction model, wherein the second analysis result is used to indicate the market development trend of the target industry;

[0008] Analyzing the customer behavior data of the target enterprise using a pre-trained customer behavior analysis model, and outputting a third analysis result by the customer behavior analysis model, wherein the third analysis result is used to indicate the asset liquidity demand of the target enterprise's customers for the target enterprise;

[0009] An asset management strategy for the target enterprise is generated based on the first analysis result, the second analysis result, and the third analysis result.

[0010] In some embodiments, the asset flow prediction model is trained as follows:

[0011] Obtain historical asset flow data for multiple historical enterprises;

[0012] Preprocessing the historical asset flow data of the plurality of historical enterprises to obtain a multidimensional tensor;

[0013] The initial time series model is trained based on the multidimensional tensor, so that the initial time series model learns the law of asset liquidity changing over time, and obtains the asset liquidity prediction model.

[0014] In some embodiments, the training process of the market volatility prediction model is as follows:

[0015] Obtain historical market data for the target industry;

[0016] Extract features from the historical market data of the target industry through a macroeconomic model to obtain market indicator features;

[0017] The initial deep learning model is trained based on the market indicator characteristics to obtain the pre-trained market fluctuation prediction model.

[0018] In some embodiments, the training process of the customer behavior analysis model is as follows:

[0019] Obtain historical customer behavior data of the target enterprise;

[0020] The initial supervised learning model is trained based on the customer historical behavior data to obtain the pre-trained customer behavior analysis model.

[0021] In some embodiments, obtaining the target enterprise's asset management strategy based on the first analysis result, the second analysis result, and the third analysis result includes:

[0022] The first analysis result, the second analysis result, and the third analysis result are input into a pre-trained asset management strategy prediction model, and the asset management strategy prediction model outputs the asset management strategy of the target enterprise.

[0023] In some embodiments, after generating the asset management strategy of the target enterprise based on the first analysis result, the second analysis result, and the third analysis result, the method further includes:

[0024] Obtaining an allocation strategy for the target enterprise's asset resources in the target enterprise's asset management strategy;

[0025] The asset resources of the target enterprise are adjusted according to the asset resource allocation strategy of the target enterprise.

[0026] In some embodiments, adjusting the target enterprise's assets and resources according to the target enterprise's asset and resource allocation strategy includes any of the following:

[0027] If a first adjustment value for adjusting the asset resources of the target enterprise is less than a first preset threshold value and the first adjustment value is greater than or equal to a second preset threshold value, evaluating the asset management strategy; and if the asset management strategy passes the evaluation, adopting the asset management strategy to manage the assets of the target enterprise, wherein the first preset threshold value is greater than the second preset threshold value;

[0028] When the first adjustment value for adjusting the asset resources of the target enterprise is greater than or equal to the first preset threshold, the first adjustment value is divided into multiple second adjustment values, a new asset management strategy is generated based on the multiple second adjustment values, and the new asset management strategy is adopted to manage the assets of the target enterprise.

[0029] To achieve the above-mentioned purpose, a second aspect of an embodiment of the present application provides a device for generating an enterprise asset management policy, the device comprising:

[0030] A data acquisition module, configured to acquire asset flow data of a target enterprise, customer behavior data of the target enterprise, and market data of a target industry, where the target industry is the industry to which the target enterprise belongs;

[0031] a first analysis module, configured to analyze the asset flow data of the target enterprise using a pre-trained asset flow prediction model, and output a first analysis result from the asset flow prediction model, wherein the first analysis result is used to indicate the asset liquidity of the target enterprise;

[0032] a second analysis module, configured to analyze the market data of the target industry using a pre-trained market fluctuation prediction model, and output a second analysis result from the market fluctuation prediction model, wherein the second analysis result is used to indicate the market development trend of the target industry;

[0033] a third analysis module, configured to analyze the customer behavior data of the target enterprise using a pre-trained customer behavior analysis model, and output a third analysis result from the customer behavior analysis model, wherein the third analysis result is used to indicate the asset liquidity demand of the target enterprise's customers for the target enterprise;

[0034] A strategy generation module is used to generate an asset management strategy for the target enterprise based on the first analysis result, the second analysis result, and the third analysis result.

[0035] To achieve the above-mentioned purpose, the third aspect of an embodiment of the present application proposes an electronic device, which includes a memory and a processor, the memory stores a computer program, and the processor implements the method for generating the enterprise asset management strategy described in the first aspect when executing the computer program.

[0036] To achieve the above-mentioned purpose, the fourth aspect of an embodiment of the present application proposes a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the method for generating an enterprise asset management strategy described in the first aspect above.

[0037] To achieve the above-mentioned purpose, an embodiment of the present application may provide a computer program product for implementation. When the instructions in the computer program product are executed by a processor of an electronic device, the electronic device implements the method for generating an enterprise asset management strategy described in the first aspect above.

[0038] The method, device, electronic device, and medium for generating an enterprise asset management strategy proposed in this application improve the rationality of an enterprise's asset allocation strategy through multi-dimensional data analysis and model prediction. Specifically, first, asset flow prediction models are used to analyze asset flow data to accurately grasp the enterprise's asset liquidity status and enable the enterprise to clearly understand its own capital turnover capacity. Second, market fluctuation prediction models are used to analyze industry market data and grasp market development trends, allowing enterprises to plan ahead and adapt to market changes. Third, customer behavior analysis models provide insight into customer demand for asset liquidity, helping enterprises better meet customer expectations. Finally, the results of these three analyses are combined to formulate an asset management strategy to further improve the rationality of the enterprise's asset allocation strategy. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 This is a flow chart of a method for generating an enterprise asset management strategy provided by an embodiment of the present application;

[0040] Figure 2 yes Figure 1 A flowchart of the asset flow prediction model training process in step S102 in FIG.

[0041] Figure 3 yes Figure 1 A flow chart of the training process of the market volatility prediction model in step S103;

[0042] Figure 4 yes Figure 1A flowchart of the training process of the customer behavior analysis model in step S104;

[0043] Figure 5 This is a schematic diagram of the structure of a device for generating an enterprise asset management strategy provided by an embodiment of the present application;

[0044] Figure 6 This is a schematic diagram of the hardware structure of the electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0046] It should be noted that although the device schematics illustrate functional module divisions and the flowcharts illustrate logical sequences, in certain circumstances, the steps shown or described may be performed in a sequence that differs from the module divisions in the device or the sequence in the flowcharts. The terms "first," "second," and so on, in the specification, claims, and drawings, are used to distinguish similar items and are not necessarily used to describe a specific sequence or precedence.

[0047] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application pertains. The terms used herein are for the purpose of describing the embodiments of this application only and are not intended to limit this application.

[0048] A company's asset management is a core component of asset-liability management and directly impacts its solvency and market competitiveness. A company's asset allocation strategy, a crucial foundation for asset management, can optimize its asset allocation, thereby enhancing its solvency and market competitiveness. However, related technologies primarily rely on historical data and simple statistical models, supplemented by manual experience, to formulate asset allocation strategies, resulting in poor rationality.

[0049] Based on this, the embodiments of the present application provide a method, device, electronic device and medium for generating an enterprise asset management strategy, aiming to solve the problem that the enterprise asset allocation strategy formulated by the existing enterprise asset management strategy generation method is less rational.

[0050] The method, device, electronic device and medium for generating an enterprise asset management policy provided in the embodiments of the present application are specifically illustrated through the following embodiments. First, the method for generating an enterprise asset management policy in the embodiments of the present application is described.

[0051] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Artificial Intelligence (AI) is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use knowledge to achieve optimal results.

[0052] Fundamental AI technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interaction systems, and mechatronics. AI software technologies primarily encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0053] The method for generating an enterprise asset management strategy provided in the embodiment of the present application relates to the field of financial technology. The method for generating an enterprise asset management strategy provided in the embodiment of the present application can be applied to a terminal, can be applied to a server side, or can be software running in a terminal or a server side. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc.; the server side can be configured as an independent physical server, or as a server cluster or distributed system composed of multiple physical servers, or as a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application that implements the method for generating an enterprise asset management strategy, etc., but is not limited to the above forms.

[0054] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and the like. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, and the like that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments in which tasks are performed by remote processing devices connected via a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media, including storage devices.

[0055] It should be noted that in each specific embodiment of the present application, when it comes to the need to perform relevant processing based on data related to the user's identity or characteristics, such as user information, user behavior data, user historical data, and user location information, the user's permission or consent will be obtained first, and the collection, use, and processing of such data will comply with relevant laws, regulations, and standards. In addition, when the embodiment of the present application needs to obtain the user's sensitive personal information, the user's separate permission or consent will be obtained through a pop-up window or by jumping to a confirmation page. After clearly obtaining the user's separate permission or consent, the necessary user-related data for the normal operation of the embodiment of the present application will be obtained.

[0056] Figure 1 This is a flow chart of the method for generating an enterprise asset management strategy provided by the embodiment of the present application. Figure 1 The method for generating an enterprise asset management strategy provided in the embodiment of the present application may include but is not limited to steps S101 to S105.

[0057] Step S101: Acquire asset flow data of a target enterprise, customer behavior data of the target enterprise, and market data of a target industry, where the target industry is the industry to which the target enterprise belongs.

[0058] In this step, the target enterprise can be a financial insurance enterprise, a medical and health consulting enterprise, a retirement management service-related enterprise, a physical sales enterprise, or a marketing enterprise, without limitation here.

[0059] Asset flow data refers to the inflow and outflow of corporate asset resources (such as cash, investment portfolios, premiums receivable, reserves, etc.), which can be obtained by integrating data such as corporate financial statements and bank transaction data.

[0060] The customer behavior data of the target enterprise refers to the interactive behavior of the target enterprise's customers towards the target enterprise, such as purchasing behavior of the target enterprise's products, claims behavior against the target enterprise, etc., which can be obtained through corporate reports, social media and other channels.

[0061] The market data of the target industry refers to the macro changes and micro data of the target industry, such as market size, market growth rate, market share of major players and product structure, which can be obtained through macroeconomic databases, industry association reports, market research reports, etc.

[0062] For example, in the financial insurance industry, the target company is Insurance Company A. Asset flow data is obtained from the annual financial report released by Company A, customer behavior data of Company A is obtained from the internal data of Company A, and market data of the insurance industry is obtained from industry statistics, reports and white papers released by the Insurance Industry Association.

[0063] Step S102: Analyze the asset flow data of the target enterprise using a pre-trained asset flow prediction model, and output a first analysis result from the asset flow prediction model, where the first analysis result is used to indicate the asset liquidity of the target enterprise.

[0064] In this step, the pre-trained asset flow prediction model is a pre-trained neural network model. This model is used to predict the future direction, speed, and scale of the company's asset flows. The neural network model can be a time series model, a machine learning model, or a deep learning model, without limitation.

[0065] The first analysis result is a quantitative analysis result of the enterprise's asset liquidity. For example, it can be a score of the enterprise's asset liquidity, a predicted value of the enterprise's asset liquidity ratio (cash ratio, quick ratio, etc.), or a judgment on the enterprise's asset liquidity trend, which is not limited here.

[0066] For example, in the financial insurance industry, for Insurance Company A, quarterly balance sheets, cash flow statements, portfolio details, and relevant market interest rate data for the past five years are obtained. This data is used as Company A's current asset data and fed into a pre-trained asset liquidity prediction model. The model then outputs a first analysis result, which can be a liquidity score for Company A, representing its asset liquidity risk. It can also be a forecast of key indicators, such as the company's cash ratio (cash and cash equivalents / current liabilities). It can also be a trend forecast and early warning for Company A's asset liquidity, such as a forecast that Company A's asset liquidity will show a slight deterioration over the next six months, primarily due to a concentrated maturities of short-term bonds and slowing premium growth. It is recommended to focus on liquidity indicators at the end of the next two quarters.

[0067] Step S103: Analyze the market data of the target industry using a pre-trained market fluctuation prediction model, and output a second analysis result from the market fluctuation prediction model, where the second analysis result is used to indicate the market development trend of the target industry.

[0068] In this step, the pre-trained market volatility prediction model is learned and optimized using a large amount of historical market data to identify the patterns and driving factors of market fluctuations and the relationship between the driving factors and future market trends.

[0069] The second analysis result is a forecast of the future state of the market, which may include predicted industry growth rate, market volatility indicators, market trend direction, etc.

[0070] For example, in the financial insurance industry, for insurance company A, the total premium income of the insurance industry in the past five years, the stock price performance and index of major listed insurance companies, interest rate levels, macroeconomic indicators (such as GDP growth rate, unemployment rate) and other data are collected to form the market data of the insurance industry. The market data of the insurance industry is input into the pre-trained market volatility prediction model, and the market volatility prediction model outputs the second analysis result, wherein the second analysis result can be a trend index of the insurance industry, for example, the insurance industry trend index is expected to be +15 in the next six months, indicating that the insurance industry market tends to be moderately optimistic, or it can be the trend direction and probability of the insurance industry, for example, in the next quarter, the total premium income of the insurance industry is expected to increase with a probability of 10%, remain flat or fluctuate slightly with a probability of 80%, and decrease with a probability of 10%, etc., and can also include multiple index information.

[0071] Step S104: Analyze the customer behavior data of the target enterprise through a pre-trained customer behavior analysis model, and output a third analysis result from the customer behavior analysis model. The third analysis result is used to indicate the asset liquidity demand of the target enterprise's customers for the target enterprise.

[0072] In this step, the pre-trained customer behavior analysis model is used to process and analyze customer interaction, transaction, and feedback behavior data to understand customer behavior patterns and predict future customer behavior.

[0073] The third analysis result is used to indicate the client's liquidity demand for the target enterprise.

[0074] For example, in the financial insurance industry, for insurance company A, the online activity data of company A's policyholders (frequency of browsing surrender terms, frequency of checking policy cash values, frequency of searching for other investment products, etc.), customer service interaction data (number of customers who consulted about the surrender process, number of customers who inquired about policy loan amounts, number of inquiries expressing concerns about the economic outlook and mentioning policies, etc.), account transaction data (frequency and amount of applications for small policy loans, frequency and amount of applications for partial withdrawal of policy cash values, etc.) and other data constitute the customer behavior data of company A. The customer behavior data of company A is input into a pre-trained customer behavior analysis model, and the customer behavior analysis model outputs a third analysis result, where the third analysis result can be "According to recent customer behavior analysis, company A faces significant pressure on customer asset liquidity demand. The main manifestations are: 1. Potential surrender intention has increased (the number of customers browsing surrender terms has surged); 2. Short-term capital turnover demand has increased (the number of policy loan inquiries has increased significantly); 3. Customer risk appetite has decreased, and they seek higher liquidity allocation (increased interest in short-term financial management)."

[0075] Step S105: Generate an asset management strategy for the target enterprise based on the first analysis result, the second analysis result, and the third analysis result.

[0076] In this step, the first analysis result, the second analysis result, and the third analysis result are converted into an asset management strategy that can be executed by the enterprise. This can be achieved through a large model or expert manual judgment, which is not limited here.

[0077] For example, the first analysis result is an asset liquidity score of 65 points (lower-middle), and the cash reserves can cover operations and expected customer withdrawals in the next month, but are slightly tight, and the proportion of long-term investments is too high; the second analysis result is that the market may enter a period of adjustment, interest rates will decline, stock market volatility will increase, and the overall premium income growth rate may slow down; the third analysis result is that customers' potential demand for short-term funds has increased, which is manifested in an increase in policy loan inquiries and some customers begin to pay attention to surrender clauses; based on the above analysis results, the following asset management strategy can be generated by the big model or corporate experts. In the next quarter, some long-term fixed-income assets (such as long-term bonds) will be gradually liquidated or sold, the proportion of cash and money market funds will be increased, and the liquidity buffer will be ensured to cover at least the expected outflows in the next three months. New investment in long-term, low-liquidity projects will be suspended, stock holdings will be carefully evaluated, and consideration will be given to appropriately reducing holdings of some highly volatile stocks in exchange for more stable assets.

[0078] This implementation improves the rationality of a company's asset allocation strategy through multi-dimensional data analysis and model prediction. Specifically, first, the asset flow prediction model analyzes asset flow data to accurately grasp the company's asset liquidity status and provide a clear understanding of its capital turnover capacity. Second, the market volatility prediction model analyzes industry market data to grasp market development trends, allowing companies to plan ahead and adapt to market changes. Third, the customer behavior analysis model provides insight into customer demand for asset liquidity, helping companies better meet customer expectations. Finally, the results of these three analyses are combined to formulate an asset management strategy to further improve the rationality of the company's asset allocation strategy.

[0079] In some embodiments, as Figure 2 As shown, the training process of the asset flow prediction model in step S102 may include but is not limited to steps S201 to S203.

[0080] Step S201: Obtain historical asset flow data of multiple historical enterprises.

[0081] Step S202: pre-process the historical asset flow data of the multiple historical enterprises to obtain a multi-dimensional tensor.

[0082] Step S203: training the initial time series model based on the multidimensional tensor, so that the initial time series model learns the law of asset liquidity changing over time, and obtains the asset liquidity prediction model.

[0083] In this implementation, historical asset flow data refers to the records of capital inflows and outflows generated by an enterprise during its operations. Specifically, it can be implemented by integrating corporate financial statements and bank transaction data to provide multi-dimensional time series samples for initial time series model training.

[0084] Multidimensional tensor refers to the structured processing of asset flow data of different enterprises according to time steps, asset categories, and enterprise size dimensions. It can be achieved through data standardization, normalization and three-dimensional matrix conversion to preserve the temporal and spatial correlation between data.

[0085] A time series model refers to an algorithmic architecture that can capture the patterns of data changes over time. Specifically, it can be implemented using a long short-term memory network or a temporal convolutional network to learn the cyclical and trend characteristics of asset flows from historical asset flow data.

[0086] Specifically, during the training of the initial time series model, the asset flow data of multiple historical companies are uniformly collected, and then missing values ​​are filled and outliers are removed to ensure data integrity. Subsequently, the data of different companies are sliced ​​according to the preset time window. Each slice contains asset flow indicators with a fixed time step, such as monthly cash flow data for the past 12 months. These sliced ​​data are converted into a three-dimensional tensor structure, where the first dimension represents different company samples, the second dimension represents the time step, and the third dimension represents the asset category characteristics.

[0087] During training, the hidden layer of the initial time series model extracts features from the tensor data and optimizes the weight parameters through backpropagation, so that the initial time series model can capture the correlation between the asset flow status at different time nodes, and finally output an asset flow prediction model with predictive capabilities.

[0088] In this implementation, by constructing a multi-dimensional tensor to integrate multi-source heterogeneous data and using a deep learning model to mine implicit correlations across time and across enterprises, it is possible to more comprehensively capture the complex patterns of asset flows, thereby improving the generalization ability and time sensitivity of the asset liquidity prediction model, and providing a more accurate prediction basis for the subsequent generation of asset management strategies.

[0089] In some embodiments, as Figure 3 As shown, the training process of the market fluctuation prediction model in step S103 may include but is not limited to steps S301 to S303.

[0090] Step S301: Acquire historical market data of the target industry.

[0091] Step S302: extracting features from the historical market data of the target industry using a macroeconomic model to obtain market indicator features.

[0092] Step S303: training the initial deep learning model based on the market indicator characteristics to obtain the pre-trained market fluctuation prediction model.

[0093] In this implementation, the macroeconomic model refers to a mathematical model used to analyze the industry economic environment. Specifically, it can be implemented using a vector autoregression model or a dynamic stochastic general equilibrium model to extract key economic indicators related to industry fluctuations from historical market data.

[0094] Market indicator characteristics refer to quantitative data that reflect market supply and demand relationships, price indices, and industry cycles. Specifically, dimensionality reduction can be achieved through principal component analysis or factor analysis to convert complex market data into feature vectors that can be recognized by deep learning models.

[0095] The initial deep learning model refers to the neural network architecture used for time series forecasting. Specifically, it can be implemented using a long short-term memory network or a Transformer structure (a neural network architecture based on a self-attention mechanism) to capture market development trends by learning the time dependency of market indicator characteristics.

[0096] Specifically, during the initial deep learning model training process, historical market data for the target industry is first screened for macroeconomic indicators, such as raw material prices, industry output value, or policy change records. The macroeconomic model then extracts features from this data, generating feature vectors that capture industry cyclical fluctuations and the state of market supply and demand equilibrium. These feature vectors are then fed into the initial deep learning model for training. The model adjusts parameters through backpropagation and learns the correlations between different market indicators and future market trends. Once trained, the market volatility forecast can predict the market development direction of the target industry based on current market data.

[0097] In this implementation, by introducing a macroeconomic model for feature extraction, it is possible to filter out irrelevant data interference, and explicitly convert the driving factors of market fluctuations into interpretable feature indicators, so that the deep learning model can be more focused on learning key influencing factors, thereby improving the market volatility prediction model's ability to capture industry trends, providing a more accurate basis for judging the market development direction for the formulation of asset management strategies, and avoiding asset allocation imbalances caused by misjudgment of market trends.

[0098] In some embodiments, as Figure 4As shown, the training process of the customer behavior analysis model in step S104 may include but is not limited to steps S401 to S402.

[0099] Step S401: Obtain the target enterprise's customer historical behavior data.

[0100] Step S402: training the initial supervised learning model based on the customer historical behavior data to obtain the pre-trained customer behavior analysis model.

[0101] In this implementation, customer historical behavior data refers to the behavioral record data on transaction frequency, capital flow preference, and product selection tendency generated by corporate customers within a historical period. Specifically, it can be collected by capturing database logs and calling API interfaces, and a structured data set is formed by associating timestamp tags with customer IDs.

[0102] The initial supervised learning model refers to a machine learning algorithm framework with classification or regression functions. It can be implemented using random forest, gradient boosting decision tree or logistic regression algorithms. During the training phase, a predictive mapping relationship is established by inputting annotated customer behavior feature vectors and corresponding asset demand labels.

[0103] Specifically, the process of building a customer behavior analysis model consists of two phases: data preparation and model training. The data preparation phase extracts historical customer behavior data from enterprise systems. After filling missing values ​​and filtering outliers, the customer behavior sequence is converted into a feature vector that reflects consumption cycles, amount fluctuations, and service type preferences. The model training phase uses the feature vector and manually labeled customer liquidity demand levels as supervisory signals. Model parameters are optimized through cross-validation, ultimately resulting in a customer behavior analysis model that can predict customer demand for enterprise asset liquidity based on their behavioral characteristics.

[0104] In this implementation, a supervised learning framework is used to automatically explore the potential correlation between customer behavior patterns and asset demand. Through iterative algorithm optimization, the prediction results have a quantifiable accuracy standard, while reducing manual intervention to accurately identify the capital usage patterns of different customer groups and provide data support for enterprises to formulate differentiated current asset allocation plans.

[0105] In some implementations, obtaining the asset management strategy of the target enterprise based on the first analysis result, the second analysis result, and the third analysis result in step S105 may include, but is not limited to, step S501.

[0106] Step S501: input the first analysis result, the second analysis result, and the third analysis result into a pre-trained asset management strategy prediction model, and the asset management strategy prediction model outputs the asset management strategy of the target enterprise.

[0107] In this implementation, the asset management strategy prediction model can be an intelligent agent based on reinforcement learning to dynamically generate asset allocation plans based on multi-dimensional data.

[0108] The training process for the intelligent agent is as follows: First, a dynamic market environment simulator is constructed to simulate the impact of macroeconomic changes, market fluctuations, and customer behavior on asset allocation. This environment defines and provides a state space (including market indicators, asset portfolio characteristics, etc.) and an action space (such as the adjustment ratio of various assets). Then, reinforcement learning algorithms such as deep Q networks and policy gradient methods are used to allow the intelligent agent to learn through trial and error in the simulated environment, learning to maximize long-term rewards (such as a comprehensive measure of return and risk) through interaction. An experience replay mechanism is also introduced during training to improve learning efficiency and stability. Finally, the trained intelligent agent is verified through historical data and a simulated environment to comprehensively evaluate its decision-making ability, strategy stability, and actual return performance under different market conditions.

[0109] An asset management strategy can be expressed as a structured decision vector that includes asset class allocation ratios, liquidity thresholds, and risk hedging recommendations.

[0110] In this implementation, multi-dimensional data collaborative analysis is achieved through the pre-trained asset management strategy prediction model, which can automatically identify the implicit correlation between different factors to improve the rationality of the generated asset management strategy.

[0111] In some embodiments, after obtaining the target enterprise's asset management strategy based on the first analysis result, the second analysis result, and the third analysis result in step S105, the method for generating the enterprise's asset management strategy may also include but is not limited to steps S601 to S602.

[0112] Step S601: Obtain the configuration strategy for the target enterprise's asset resources in the target enterprise's asset management strategy.

[0113] Step S601: Adjust the asset resources of the target enterprise according to the asset resource allocation strategy of the target enterprise.

[0114] In this implementation, an allocation strategy refers to the adjustment rules set within an asset management strategy for different asset types. Specifically, this can be implemented using preset asset allocation ratios or dynamic adjustment thresholds. For example, an allocation strategy might require that cash assets maintain a minimum ratio of 20% of total assets, with an adjustment mechanism triggered when the actual ratio falls below this threshold.

[0115] Specifically, after the asset management strategy is generated, resource allocation parameters are extracted by parsing the strategy file, such as increasing the accounts receivable turnover rate to 1.2 times the industry average; asset resource adjustments can be automatically executed through the enterprise resource management system. For example, when the allocation strategy requires an increase in working capital reserves, the system automatically allocates funds to the designated account; for major asset structure adjustments, an audit task can be generated and submitted for manual confirmation before execution; by continuously monitoring the matching degree between asset change data and allocation strategies, a closed-loop optimization mechanism is established to achieve dynamic adjustments.

[0116] In this implementation, by quantifying the configuration strategy parameters and establishing an automated execution mechanism, the adjustment process is immediately triggered when it is detected that the asset configuration deviates from the preset threshold, significantly improving the timeliness and accuracy of asset adjustments.

[0117] In some implementations, adjusting the assets and resources of the target enterprise according to the configuration strategy of the assets and resources of the target enterprise in step S602 may include any of the following steps.

[0118] Step S701: When a first adjustment value for adjusting the asset resources of the target enterprise is less than a first preset threshold value and the first adjustment value is greater than or equal to a second preset threshold value, the asset management policy is evaluated; and when the asset management policy passes the evaluation, the asset management policy is adopted to manage the assets of the target enterprise, wherein the first preset threshold value is greater than the second preset threshold value.

[0119] Step S702: When the first adjustment value for adjusting the asset resources of the target enterprise is greater than or equal to the first preset threshold, the first adjustment value is divided into multiple second adjustment values, a new asset management strategy is generated according to the multiple second adjustment values, and the new asset management strategy is adopted to manage the assets of the target enterprise.

[0120] In this implementation, the first adjustment value refers to the amount or proportion of assets that need to be adjusted in the asset resource allocation strategy. This can be achieved by calculating the difference or percentage change between the current asset allocation and the target allocation, which is used to measure the size of the adjustment.

[0121] The first preset threshold refers to the pre-set upper limit of the adjustment range, which can be set as a fixed value or a dynamic variable based on the enterprise's risk tolerance or industry standards, and is used to determine whether to trigger the phased adjustment mechanism.

[0122] The second preset threshold refers to a preset lower limit of the adjustment amplitude, which can be specifically set to a fixed value or ratio lower than the first preset threshold, and is used to distinguish the boundary conditions between small adjustments and medium adjustments.

[0123] Asset management strategy evaluation refers to the process of verifying the feasibility and risk factors of the strategy. Specifically, the expected return volatility or liquidity gap indicator after the strategy is executed can be calculated through a risk assessment model. When the indicator is lower than the set threshold, the evaluation is judged to have passed. Manual evaluation can also be used, which is not limited here.

[0124] Specifically, when the adjustment range of the asset resource allocation strategy is within the medium range, the strategy is first stress-tested using a risk quantification model. For example, if the adjustment value reaches 20% but does not exceed 50% of the company's quarterly cash flow, the impact of market fluctuations on the adjustment strategy is simulated. If the simulation results show that key financial indicators remain within a safe range, the strategy is allowed to be executed. If the adjustment range exceeds 30% of the company's annual revenue, the total adjustment amount is split into multiple quarters for phased implementation. The adjustment amount for each quarter can be set to no more than 10% of annual revenue, thereby avoiding liquidity risks caused by large one-time adjustments.

[0125] It should be noted that both the first preset threshold and the second preset threshold can be set according to actual conditions and are not limited here.

[0126] In this implementation method, while ensuring the security of the company's capital chain, it effectively balances the needs of asset optimization and allocation with risk control objectives; when encountering large-scale asset restructuring needs, the phased execution mechanism can ensure the stability of corporate operations and avoid the deterioration of financial indicators caused by a single adjustment overload.

[0127] Figure 5 This is a schematic diagram of the structure of the device for generating the enterprise asset management strategy provided by the embodiment of the present application. Figure 5 The embodiment of the present application further provides a device 800 for generating an enterprise asset management policy, which can implement the above-mentioned method for generating an enterprise asset management policy. The device 800 for generating an enterprise asset management policy includes:

[0128] The data acquisition module 801 is used to acquire the asset flow data of the target enterprise, the customer behavior data of the target enterprise, and the market data of the target industry, where the target industry is the industry to which the target enterprise belongs;

[0129] A first analysis module 802 is configured to analyze the asset flow data of the target enterprise using a pre-trained asset flow prediction model, and output a first analysis result from the asset flow prediction model, wherein the first analysis result is used to indicate the asset liquidity of the target enterprise;

[0130] A second analysis module 803 is configured to analyze the market data of the target industry using a pre-trained market fluctuation prediction model, and output a second analysis result from the market fluctuation prediction model, where the second analysis result is used to indicate the market development trend of the target industry;

[0131] A third analysis module 804 is configured to analyze the customer behavior data of the target enterprise using a pre-trained customer behavior analysis model, and output a third analysis result from the customer behavior analysis model, wherein the third analysis result is used to indicate the asset liquidity demand of the target enterprise's customers for the target enterprise;

[0132] The strategy generation module 805 is configured to generate an asset management strategy for the target enterprise based on the first analysis result, the second analysis result, and the third analysis result.

[0133] In some embodiments, the asset flow prediction model is trained as follows:

[0134] Obtain historical asset flow data for multiple historical enterprises;

[0135] Preprocessing the historical asset flow data of the plurality of historical enterprises to obtain a multidimensional tensor;

[0136] The initial time series model is trained based on the multidimensional tensor, so that the initial time series model learns the law of asset liquidity changing over time, and obtains the asset liquidity prediction model.

[0137] In some embodiments, the training process of the market volatility prediction model is as follows:

[0138] Obtain historical market data for the target industry;

[0139] Extract features from the historical market data of the target industry through a macroeconomic model to obtain market indicator features;

[0140] The initial deep learning model is trained based on the market indicator characteristics to obtain the pre-trained market fluctuation prediction model.

[0141] In some embodiments, the training process of the customer behavior analysis model is as follows:

[0142] Obtain historical customer behavior data of the target enterprise;

[0143] The initial supervised learning model is trained based on the customer historical behavior data to obtain the pre-trained customer behavior analysis model.

[0144] In some implementations, the policy generation module 805 includes:

[0145] The strategy generation submodule is used to input the first analysis result, the second analysis result and the third analysis result into a pre-trained asset management strategy prediction model, and the asset management strategy prediction model outputs the asset management strategy of the target enterprise.

[0146] In some implementations, the enterprise asset management policy generation device 800 further includes:

[0147] A strategy acquisition module, configured to acquire a configuration strategy for the target enterprise's asset resources in the target enterprise's asset management strategy;

[0148] The asset adjustment module is used to adjust the asset resources of the target enterprise according to the configuration strategy of the asset resources of the target enterprise.

[0149] In some embodiments, the asset adjustment module includes:

[0150] a first adjustment submodule, configured to evaluate the asset management strategy if a first adjustment value for adjusting the asset resources of the target enterprise is less than a first preset threshold and the first adjustment value is greater than or equal to a second preset threshold, and to adopt the asset management strategy to manage the assets of the target enterprise if the asset management strategy passes the evaluation, wherein the first preset threshold is greater than the second preset threshold;

[0151] The second adjustment submodule is used to divide the first adjustment value for adjusting the asset resources of the target enterprise into multiple second adjustment values ​​when the first adjustment value is greater than or equal to the first preset threshold, generate a new asset management strategy based on the multiple second adjustment values, and adopt the new asset management strategy to manage the assets of the target enterprise.

[0152] The specific implementation of the device 800 for generating an enterprise asset management policy is substantially the same as the specific embodiment of the method for generating an enterprise asset management policy described above, and will not be described in detail herein.

[0153] An embodiment of the present application further provides an electronic device comprising a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the aforementioned method for generating an enterprise asset management policy. The electronic device can be any intelligent terminal, including a desktop computer, a tablet computer, a mobile phone, and an in-vehicle computer.

[0154] See also Figure 6 , Figure 6 This is a schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present application. The electronic device includes:

[0155] The processor 901 can be implemented as a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application.

[0156] The memory 902 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 902 can store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 902 and is called by the processor 901 to execute the method for generating an enterprise asset management policy in the embodiments of this application.

[0157] Input / output interface 903, used to implement information input and output;

[0158] Communication interface 904, used to implement communication interaction between this device and other devices, which can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WiFi, Bluetooth, etc.);

[0159] Bus 905 , which transmits information between various components of the device (e.g., processor 901 , memory 902 , input / output interface 903 , and communication interface 904 );

[0160] The processor 901 , the memory 902 , the input / output interface 903 and the communication interface 904 are connected to each other in communication within the device via a bus 905 .

[0161] An embodiment of the present application further provides a computer-readable storage medium storing a computer program, which implements the above-mentioned method for generating an enterprise asset management strategy when executed by a processor.

[0162] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0163] In addition, the embodiments of the present application may be implemented by providing a computer program product. When the instructions in the computer program product are executed by a processor of an electronic device, the electronic device implements the method for generating an enterprise asset management policy in the above embodiment.

[0164] The method, device, electronic device and medium for generating enterprise asset management strategies provided in the embodiments of the present application improve the rationality of enterprise asset allocation strategies through multi-dimensional data analysis and model prediction. Specifically, first, asset flow prediction models are used to analyze asset flow data to accurately grasp the liquidity status of enterprise assets and enable enterprises to clearly understand their own capital turnover capabilities. Secondly, market fluctuation prediction models are used to analyze industry market data and grasp market development trends so that enterprises can plan ahead and adapt to market changes. Thirdly, customer behavior analysis models provide insights into customer demand for asset liquidity and help enterprises better meet customer expectations. Finally, the results of the three analyses are combined to formulate asset management strategies to further improve the rationality of enterprise asset allocation strategies.

[0165] The embodiments described in the embodiments of this application are intended to more clearly illustrate the technical solutions of the embodiments of this application and do not constitute a limitation on the technical solutions provided by the embodiments of this application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0166] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0167] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, i.e., they may be located in one place or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of this embodiment.

[0168] Those skilled in the art will appreciate that all or some of the steps in the methods, systems, and functional modules / units in the devices disclosed above may be implemented as software, firmware, hardware, or appropriate combinations thereof.

[0169] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0170] It should be understood that in this application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following items" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0171] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the above-mentioned units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0172] The units described above as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0173] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0174] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes multiple instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: various media that can store programs, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0175] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but are not intended to limit the scope of the present invention. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and essence of the present invention should be within the scope of the present invention.

Claims

1. A method for generating an enterprise asset management strategy, characterized in that: The method comprises: Obtaining asset flow data of a target enterprise, customer behavior data of the target enterprise, and market data of a target industry, where the target industry is the industry to which the target enterprise belongs; Analyzing the asset flow data of the target enterprise using a pre-trained asset flow prediction model, and outputting a first analysis result by the asset flow prediction model, wherein the first analysis result is used to indicate the asset liquidity of the target enterprise; Analyzing the market data of the target industry using a pre-trained market fluctuation prediction model, and outputting a second analysis result by the market fluctuation prediction model, wherein the second analysis result is used to indicate the market development trend of the target industry; Analyzing the customer behavior data of the target enterprise using a pre-trained customer behavior analysis model, and outputting a third analysis result by the customer behavior analysis model, wherein the third analysis result is used to indicate the asset liquidity demand of the target enterprise's customers for the target enterprise; An asset management strategy for the target enterprise is generated based on the first analysis result, the second analysis result, and the third analysis result.

2. The method according to claim 1, characterized in that The training process of the asset flow prediction model is as follows: Obtain historical asset flow data for multiple historical enterprises; Preprocessing the historical asset flow data of the plurality of historical enterprises to obtain a multidimensional tensor; The initial time series model is trained based on the multidimensional tensor, so that the initial time series model learns the law of asset liquidity changing over time, and obtains the asset liquidity prediction model.

3. The method according to claim 1, characterized in that The training process of the market volatility prediction model is as follows: Obtain historical market data for the target industry; Extract features from the historical market data of the target industry through a macroeconomic model to obtain market indicator features; The initial deep learning model is trained based on the market indicator characteristics to obtain the pre-trained market fluctuation prediction model.

4. The method according to claim 1, wherein The training process of the customer behavior analysis model is as follows: Obtain historical customer behavior data of the target enterprise; The initial supervised learning model is trained based on the customer historical behavior data to obtain the pre-trained customer behavior analysis model.

5. The method according to claim 1, wherein The obtaining of the target enterprise's asset management strategy based on the first analysis result, the second analysis result, and the third analysis result includes: The first analysis result, the second analysis result, and the third analysis result are input into a pre-trained asset management strategy prediction model, and the asset management strategy prediction model outputs the asset management strategy of the target enterprise.

6. The method according to claim 1, after generating the asset management strategy of the target enterprise based on the first analysis result, the second analysis result, and the third analysis result, the method further comprises: Obtaining a configuration strategy for the target enterprise's asset resources in the target enterprise's asset management strategy; The asset resources of the target enterprise are adjusted according to the asset resource allocation strategy of the target enterprise.

7. The method according to claim 6, wherein adjusting the assets and resources of the target enterprise according to the asset and resource allocation strategy of the target enterprise comprises any one of the following: When a first adjustment value of the asset resources of the target enterprise is less than a first preset threshold value and the first adjustment value is greater than or equal to a second preset threshold value, the asset management strategy is evaluated. When the asset management strategy passes the evaluation, the asset management strategy is adopted to manage the assets of the target enterprise, wherein: The first preset threshold is greater than the second preset threshold; When the first adjustment value for adjusting the asset resources of the target enterprise is greater than or equal to the first preset threshold, the first adjustment value is divided into multiple second adjustment values, a new asset management strategy is generated based on the multiple second adjustment values, and the new asset management strategy is adopted to manage the assets of the target enterprise.

8. A device for generating an enterprise asset management strategy, characterized in that: The device comprises: A data acquisition module, configured to acquire asset flow data of a target enterprise, customer behavior data of the target enterprise, and market data of a target industry, where the target industry is the industry to which the target enterprise belongs; a first analysis module, configured to analyze the asset flow data of the target enterprise using a pre-trained asset flow prediction model, and output a first analysis result from the asset flow prediction model, wherein the first analysis result is used to indicate the asset liquidity of the target enterprise; a second analysis module, configured to analyze the market data of the target industry using a pre-trained market fluctuation prediction model, and output a second analysis result from the market fluctuation prediction model, wherein the second analysis result is used to indicate the market development trend of the target industry; a third analysis module, configured to analyze the customer behavior data of the target enterprise using a pre-trained customer behavior analysis model, and output a third analysis result from the customer behavior analysis model, wherein the third analysis result is used to indicate the asset liquidity demand of the target enterprise's customers for the target enterprise; A strategy generation module is used to generate an asset management strategy for the target enterprise based on the first analysis result, the second analysis result, and the third analysis result.

9. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the method for generating an enterprise asset management strategy according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for generating an enterprise asset management policy according to any one of claims 1 to 7 is implemented.

Images