Financial asset configuration method and system in bank insurance field based on intelligent analysis

Abstract

The invention belongs to the technical field of financial asset configuration, and discloses a bank insurance field financial asset configuration method and system based on intelligent analysis, and the method comprises the steps: obtaining historical transaction data, carrying out the preprocessing of the historical transaction data, carrying out the transaction behavior evaluation through an entropy method, and obtaining behavior evaluation data; pre-processing pre-acquired asset data, and performing asset attribute dynamic adjustment by using the dynamic asset classification model to obtain asset attribute data; performing asset allocation optimization by using a multi-objective optimization algorithm to generate an optimized asset allocation scheme; and predicting a market trend by using a machine learning model according to pre-acquired market economic data, extracting market sensitivity by using a capital asset pricing model, and generating a final asset allocation scheme based on the market sensitivity. According to the invention, through the behavior evaluation data and the asset attribute data, a traditional insurance product marketing thought is broken through, and combination with actual user scenarized marketing is realized through a multi-objective optimization algorithm.

Description

Technical Field

[0001] The present invention relates to the field of financial asset allocation, and in particular to a method and system for financial asset allocation in the banking and insurance field based on intelligent analysis. Background Art

[0002] In the current banking and insurance industry, financial managers face many challenges when marketing complex financial products, especially in the use of marketing tools. The marketing tools they rely on are complex and diverse, and lack unified and integrated platform support. Therefore, it is difficult to run through all aspects of product sales, resulting in inefficient use of marketing tools, which in turn increases sales difficulty and deviates from the potential real needs of bank customers. This traditional marketing method and tool cannot meet the technical needs of the banking sector for product marketing.

[0003] In particular, the actual marketing process suffers from a mismatch between customer needs and product selling points. This is primarily due to a lack of in-depth analysis and accurate identification of the personalized needs of bank customers, resulting in a disconnect between marketing content and core demands such as actual risk preferences and financial planning. Furthermore, the existing marketing tool system is overly complex and fragmented, lacking a unified, integrated operating platform. Financial managers struggle to complete the sales process for a wide range of complex financial products using a single tool. The high barrier to entry and cumbersome operation not only impact marketing efficiency but also hinder the rapid onboarding and training of new employees amidst frequent staff turnover, further complicating the promotion of complex products.

[0004] Therefore, how to provide a financial asset allocation method and system in the banking and insurance field based on intelligent analysis is an urgent problem that needs to be solved. Summary of the Invention

[0005] The embodiments of the present invention provide a financial asset allocation method and system in the banking and insurance field based on intelligent analysis to solve the above-mentioned technical problems in the prior art.

[0006] To provide a basic understanding of some aspects of the disclosed embodiments, the following is a brief summary. This summary is not intended to be a comprehensive review, identify key or essential elements, or delineate the scope of these embodiments. Its sole purpose is to present some concepts in a simplified form as a prelude to the detailed description that follows.

[0007] According to a first aspect of an embodiment of the present invention, a financial asset allocation method in the banking and insurance field based on intelligent analysis is provided.

[0008] In one embodiment, a method for allocating financial assets in the banking and insurance sector based on intelligent analysis includes:

[0009] Obtain historical transaction data, pre-process the historical transaction data, and use the entropy method to combine the pre-processed historical transaction data to evaluate transaction behavior and obtain behavior evaluation data;

[0010] Preprocess the pre-acquired asset data, and dynamically adjust the asset attributes of the real-time acquired asset data using a dynamic asset classification model based on the pre-processed asset data to obtain asset attribute data;

[0011] Based on behavioral assessment data and asset attribute data, we use multi-objective optimization algorithms to optimize asset allocation and generate optimized asset allocation plans;

[0012] Based on the pre-acquired market economic data, the machine learning model is used to predict market trends, and the capital asset pricing model is used in combination with the market trend prediction results to extract market sensitivity. Based on the market sensitivity, the optimized asset allocation plan is adjusted to generate the final asset allocation plan.

[0013] In one embodiment, historical transaction data is obtained, preprocessed, and transaction behavior evaluation is performed using an entropy method combined with the preprocessed historical transaction data. The obtained behavior evaluation data includes:

[0014] Obtain historical transaction data and use interpolation to process missing values ​​in the historical transaction data to obtain completed historical transaction data;

[0015] Historical transaction data includes user information data, historical transaction behavior data, and questionnaire data;

[0016] Perform outlier processing on the padded historical transaction data based on the interquartile range to obtain cleaned historical transaction data, and normalize the cleaned historical transaction data using the standard score to obtain pre-processed historical transaction data;

[0017] Based on the pre-processed historical transaction data, statistical analysis methods are used to extract multi-dimensional behavioral characteristic indicators, and a behavioral evaluation indicator system is constructed using the multi-dimensional behavioral characteristic indicators;

[0018] Based on the behavioral evaluation index system, the entropy method is used to calculate the weight of each indicator, and the multi-dimensional behavioral characteristic indicators are comprehensively evaluated using weighted summation according to the weight of each indicator to obtain behavioral evaluation data.

[0019] In one embodiment, based on the behavior evaluation index system, the entropy method is used to calculate the weight of each index, and the multi-dimensional behavior characteristic index is comprehensively evaluated using weighted summation according to the weight of each index. The obtained behavior evaluation data includes:

[0020] Based on the behavioral evaluation index system, the entropy method is used to calculate the proportion of each indicator, and the information entropy value of each indicator is calculated according to the proportion of each indicator;

[0021] The difference coefficient of each indicator is calculated through the information entropy value of each indicator, and the weight of each indicator is obtained by using the difference coefficient of each indicator;

[0022] According to the weight of each indicator, the multi-dimensional behavioral characteristic indicators are comprehensively evaluated using weighted summation to obtain behavioral assessment data.

[0023] In one embodiment, the pre-acquired asset data is pre-processed, and the asset attributes of the real-time acquired asset data are dynamically adjusted using a dynamic asset classification model based on the pre-processed asset data. The obtained asset attribute data includes:

[0024] Process missing values ​​and outliers on the pre-acquired asset data to obtain cleaned asset data, normalize the cleaned asset data using standard scores to obtain standard asset data, and perform time alignment on the standard asset data to obtain pre-processed asset data;

[0025] Based on the pre-processed asset data, time series modeling is used to extract asset attribute characteristics, and the importance of asset attribute characteristics is evaluated using the Shapley additive interpretation method. Based on the importance evaluation results, the asset attribute characteristics are screened to obtain key asset characteristics;

[0026] The K-means clustering algorithm is used to perform unsupervised cluster analysis on key asset features to obtain asset groups, and a dynamic asset classification model is constructed based on the asset groups. The dynamic asset classification model is used to dynamically adjust the asset attributes of the asset data obtained in real time to obtain asset attribute data.

[0027] In one embodiment, an unsupervised cluster analysis of key asset features is performed using a K-means clustering algorithm to obtain asset groups. A dynamic asset classification model is constructed based on the asset groups. The dynamic asset classification model is used to dynamically adjust the asset attributes of the real-time acquired asset data. The obtained asset attribute data includes:

[0028] Based on the key asset characteristics, the optimal number of clusters of the K-means clustering algorithm is determined by the elbow method, and unsupervised cluster analysis is performed on the key asset characteristics according to the optimal number of clusters to obtain the asset clustering results;

[0029] Generate asset group labels based on asset clustering results, train an extreme gradient boosting model based on the asset group labels, and use the trained extreme gradient boosting model to build a dynamic asset classification model;

[0030] The dynamic asset classification model is used for dynamically adjusting asset attribute data obtained in real time to obtain asset attribute data.

[0031] In one embodiment, based on the behavior evaluation data and the asset attribute data, a multi-objective optimization algorithm is used for asset allocation optimization to generate an optimized asset allocation scheme, which includes:

[0032] Based on the asset attribute data, a plurality of optimization objectives are constructed, and a linear weighted summation method is used to generate an optimization objective function according to the plurality of optimization objectives.

[0033] According to the behavior evaluation data, a user risk assessment result is obtained, and the user risk bearing capacity is judged based on the user risk assessment result, and the user risk bearing capacity is converted into a constraint condition of the multi-objective optimization algorithm.

[0034] A genetic algorithm is used to combine the optimization objective function and the constraint condition to generate a Pareto optimal solution set, and a superior-inferior solution distance method is used to perform asset allocation optimization based on the Pareto optimal solution set to generate an optimized asset allocation scheme.

[0035] In one embodiment, a genetic algorithm is used to combine the optimization objective function and the constraint condition to generate a Pareto optimal solution set, and a superior-inferior solution distance method is used to perform asset allocation optimization based on the Pareto optimal solution set to generate an optimized asset allocation scheme, which includes:

[0036] A genetic algorithm is used to perform multi-objective optimization on the optimization objective function and the constraint condition, and a Pareto optimal solution set is generated based on the multi-objective optimization result.

[0037] The Pareto optimal solution set is standardized to obtain a standardized Pareto optimal solution set, and a superior-inferior solution distance method is used to determine a positive ideal solution and a negative ideal solution according to the standardized Pareto optimal solution set.

[0038] The Euclidean distance of each Pareto optimal solution to the positive ideal solution and the negative ideal solution is calculated, and the proximity of the Pareto optimal solution to the positive ideal solution is calculated by calculating the obtained Euclidean distance.

[0039] The Pareto optimal solution with the highest proximity is selected for asset allocation optimization to generate an optimized asset allocation scheme.

[0040] In one embodiment, a machine learning model is used to predict market trends according to pre-acquired market economic data, and a capital asset pricing model is used to extract market sensitivity combined with the market trend prediction result, and the optimized asset allocation scheme is adjusted based on the market sensitivity to generate a final asset allocation scheme, which includes:

[0041] Processing missing values ​​and outliers on pre-acquired market economic data to obtain cleaned market economic data, and normalizing and time-aligning the cleaned market economic data to obtain standardized market economic data;

[0042] The standardized market economic data is divided into a training set and a test set, and the long short-term memory network model is trained using the training set. The market trend prediction is then performed on the real-time acquired market economic data based on the trained long short-term memory network model to obtain the market trend prediction results.

[0043] The capital asset pricing model is used in combination with market trend forecast results to calculate the market sensitivity of various assets. The performance expectations of various assets are judged based on the market sensitivity of various assets. Based on the performance expectations of various assets, the optimized asset allocation plan is adjusted to generate the final asset allocation plan.

[0044] In one embodiment, the capital asset pricing model is used in combination with market trend forecast results to calculate the market sensitivity of each asset class. The performance expectations of each asset class are determined based on the market sensitivity of each asset class. The optimized asset allocation plan is adjusted based on the performance expectations of each asset class. The final asset allocation plan includes:

[0045] Based on the capital asset pricing model, a single-factor linear regression analysis is conducted on the market trend forecast results, and the market sensitivity of various assets is calculated using the single-factor linear regression analysis results;

[0046] Determine the performance expectations of each asset class based on its market sensitivity, identify priority assets based on the expected performance results of each asset class, and obtain priority assets;

[0047] Based on the priority assets, the asset ratio of the optimized asset allocation plan is dynamically adjusted to generate the final asset allocation plan.

[0048] According to a second aspect of an embodiment of the present invention, a financial asset allocation system in the banking and insurance field based on intelligent analysis is provided.

[0049] In one embodiment, a financial asset allocation system for the banking and insurance sector based on intelligent analysis includes: a behavior assessment data acquisition module, an asset attribute data acquisition module, an asset allocation plan generation module, and an asset allocation plan adjustment module;

[0050] A behavior evaluation data acquisition module is used to obtain historical transaction data, pre-process the historical transaction data, and use the entropy method to combine the pre-processed historical transaction data to evaluate transaction behavior and obtain behavior evaluation data;

[0051] The asset attribute data acquisition module is used to pre-process the pre-acquired asset data and dynamically adjust the asset attributes of the real-time acquired asset data using a dynamic asset classification model based on the pre-processed asset data to obtain asset attribute data;

[0052] The asset allocation plan generation module is used to optimize asset allocation based on behavioral assessment data and asset attribute data using a multi-objective optimization algorithm to generate an optimized asset allocation plan;

[0053] The asset allocation plan adjustment module is used to use machine learning models to predict market trends based on pre-acquired market economic data, and to extract market sensitivity using the capital asset pricing model combined with market trend prediction results. Based on market sensitivity, the optimized asset allocation plan is adjusted to generate the final asset allocation plan.

[0054] According to a third aspect of embodiments of the present invention, a computer device is provided.

[0055] In some embodiments, the computer device includes a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, it implements the steps of the above-mentioned financial asset allocation method in the banking and insurance field based on intelligent analysis.

[0056] According to a fourth aspect of embodiments of the present invention, a computer-readable storage medium is provided.

[0057] In one embodiment, a computer program is stored on a computer-readable storage medium, and when the computer program is executed by a processor, the steps of the above-mentioned financial asset allocation method in the banking and insurance field based on intelligent analysis are implemented.

[0058] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0059] 1. This invention breaks through the traditional insurance product marketing ideas by using behavioral assessment data and asset attribute data, and combines it with actual user scenario marketing through a multi-objective optimization algorithm combined with market sensitivity, accurately capturing the product needs of potential bank customers and providing sales personnel with a new marketing method and solution.

[0060] 2. The present invention uses behavioral assessment data to establish customer profiles and demand analysis of bank-specific customers for the marketing of complex financial products, further improving the matching degree between the asset allocation plan of financial product marketing and customer needs; and through asset attribute data, the asset allocation plan can simultaneously meet the support of multi-type product marketing, achieving simple, centralized, efficient and precise matching; finally, through market sensitivity, the present invention can adjust the asset allocation plan with the help of macro-market economic data, thereby improving the adaptability and robustness of the asset portfolio in different market environments.

[0061] 3. The present invention uses behavioral assessment data, asset attribute data, asset allocation optimization and market sensitivity to achieve accurate customer portrait and demand analysis, optimize asset allocation and generate product recommendation plans based on customer needs, and cover the application logic ideas of complex financial products in multiple categories. It can enable practitioners to concentrate from scattered product marketing methods to a unified marketing logic, further improve the types of products held by customers, thereby increasing customer stickiness and cross-selling opportunities; it can also optimize asset allocation, encourage customers to try more products, dynamically adjust weights to achieve real-time or regular market data updates, and ensure the timeliness of recommendations.

[0062] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0064] Figure 1 This is a flowchart of a method for allocating financial assets in the banking and insurance sector based on intelligent analysis according to an exemplary embodiment;

[0065] Figure 2 This is a schematic diagram of the structure of a financial asset allocation system in the banking and insurance field based on intelligent analysis according to an exemplary embodiment;

[0066] Figure 3 is a structural diagram of a computer device according to an exemplary embodiment;

[0067] Figure 4 This is a flowchart of user login system management in a financial asset allocation system in the banking and insurance field based on intelligent analysis according to an exemplary embodiment;

[0068] Figure 5 This is a logic diagram of user login verification of a Torner engine in a financial asset allocation system in the banking and insurance field based on intelligent analysis, according to an exemplary embodiment;

[0069] Figure 6 This is a diagram illustrating a system access layer architecture in a financial asset allocation system in the banking and insurance field based on intelligent analysis according to an exemplary embodiment;

[0070] Figure 7 It is a flowchart of real-time data service and processing in a financial asset allocation system in the banking and insurance field based on intelligent analysis according to an exemplary embodiment. DETAILED DESCRIPTION

[0071] The following description and accompanying drawings sufficiently illustrate the specific embodiments herein to enable those skilled in the art to practice them. Portions and features of some embodiments may be included in or substituted for portions and features of other embodiments. The scope of the embodiments herein includes the entire scope of the claims, including all available equivalents thereof. Herein, the terms "first," "second," and the like are used solely to distinguish one element from another and do not require or imply any actual relationship or order between these elements. In practice, the first element can also be referred to as the second element, and vice versa. Furthermore, the terms "comprise," "comprising," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a structure, device, or apparatus comprising a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such structure, device, or apparatus. Without further limitation, an element defined by the phrase "comprising a..." does not preclude the presence of other identical elements in the structure, device, or apparatus comprising the element. The various embodiments herein are described in a progressive manner, with each embodiment focusing on its differences from the other embodiments. Similar or identical parts between the various embodiments can be referenced to each other.

[0072] The terms "longitudinal", "transverse", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like used herein to indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, are intended only to facilitate the description of this document and simplify the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation on the present invention. In the description herein, unless otherwise specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, they can be mechanical or electrical connections, or they can be internal connections between two elements, they can be directly connected, or they can be indirectly connected through an intermediate medium. For those of ordinary skill in the art, the specific meanings of the above terms can be understood according to the specific circumstances.

[0073] As used herein, unless otherwise specified, the term "plurality" means two or more.

[0074] In this document, the character " / " indicates that the preceding and following objects are in an "or" relationship. For example, A / B means: A or B.

[0075] In this article, the term "and / or" describes the association relationship between objects, indicating that three relationships can exist. For example, A and / or B means: A or B, or A and B.

[0076] It should be understood that, although the various steps in the flowchart are shown in sequence as indicated by the arrows, these steps are not necessarily performed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps may be performed in other orders. Moreover, at least a portion of the steps in the figure may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily performed at the same time, but may be performed at different times. The execution order of these sub-steps or stages is not necessarily to be performed in sequence, but may be performed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0077] Each module in the device or system of the present application can be implemented in whole or in part by software, hardware, or a combination thereof. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software so that the processor can call and execute the operations corresponding to the above modules.

[0078] In the absence of conflict, the embodiments of the present invention and the features thereof may be combined with each other.

[0079] Figure 1 An embodiment of the financial asset allocation method in the banking and insurance field based on intelligent analysis of the present invention is shown.

[0080] In this optional embodiment, the method for allocating financial assets in the banking and insurance sector based on intelligent analysis includes:

[0081] Step S101: Acquire historical transaction data, pre-process the historical transaction data, and use an entropy method to evaluate transaction behavior based on the pre-processed historical transaction data to obtain behavior evaluation data.

[0082] Step S102: pre-processing the pre-acquired asset data, and dynamically adjusting the asset attributes of the real-time acquired asset data using a dynamic asset classification model based on the pre-processed asset data to obtain asset attribute data;

[0083] Step S103: Based on the behavior evaluation data and the asset attribute data, a multi-objective optimization algorithm is used to optimize the asset allocation and generate an optimized asset allocation plan;

[0084] Step S104: Use a machine learning model to predict market trends based on pre-acquired market economic data, and use the capital asset pricing model combined with the market trend prediction results to extract market sensitivity. Adjust the optimized asset allocation plan based on the market sensitivity to generate a final asset allocation plan.

[0085] In the optional embodiment, historical transaction data is acquired, the historical transaction data is preprocessed, and the transaction behavior is evaluated by using the entropy method combined with the preprocessed historical transaction data to obtain behavior evaluation data, including:

[0086] The historical transaction data is acquired, and the interpolation method is used to process the missing values of the historical transaction data to obtain the historical transaction data after filling;

[0087] The historical transaction data includes user information data, historical transaction behavior data and questionnaire data;

[0088] Based on the quartile distance, the historical transaction data after filling is processed to obtain the cleaned historical transaction data, and the cleaned historical transaction data is normalized by using the standard score to obtain the preprocessed historical transaction data;

[0089] According to the preprocessed historical transaction data, multi-dimensional behavior characteristic indexes are extracted by using the statistical analysis method, and the behavior evaluation index system is constructed by using the multi-dimensional behavior characteristic indexes;

[0090] Based on the behavior evaluation index system, the entropy method is used to calculate the index weight, and the multi-dimensional behavior characteristic indexes are comprehensively evaluated by using weighted summation according to the index weight to obtain behavior evaluation data.

[0091] In the optional embodiment, based on the behavior evaluation index system, the entropy method is used to calculate the index weight, and the multi-dimensional behavior characteristic indexes are comprehensively evaluated by using weighted summation according to the index weight to obtain behavior evaluation data, including:

[0092] Based on the behavior evaluation index system, the entropy method is used to calculate the index proportion, and the index information entropy value is calculated according to the index proportion;

[0093] The index difference coefficient is calculated by the index information entropy value, and the index weight is obtained by using the index difference coefficient;

[0094] According to the index weight, the multi-dimensional behavior characteristic indexes are comprehensively evaluated by using weighted summation to obtain behavior evaluation data.

[0095] In the optional embodiment, the preprocessed asset data is acquired, and the asset attribute dynamic adjustment is performed on the real-time acquired asset data by using the dynamic asset classification model according to the preprocessed asset data to obtain asset attribute data, including:

[0096] Process missing values ​​and outliers on the pre-acquired asset data to obtain cleaned asset data, normalize the cleaned asset data using standard scores to obtain standard asset data, and perform time alignment on the standard asset data to obtain pre-processed asset data;

[0097] Based on the pre-processed asset data, time series modeling is used to extract asset attribute characteristics, and the importance of asset attribute characteristics is evaluated using the Shapley additive interpretation method. Based on the importance evaluation results, the asset attribute characteristics are screened to obtain key asset characteristics;

[0098] The K-means clustering algorithm is used to perform unsupervised cluster analysis on key asset features to obtain asset groups, and a dynamic asset classification model is constructed based on the asset groups. The dynamic asset classification model is used to dynamically adjust the asset attributes of the asset data obtained in real time to obtain asset attribute data.

[0099] In this optional embodiment, an unsupervised cluster analysis of key asset features is performed using a K-means clustering algorithm to obtain asset groups. A dynamic asset classification model is constructed based on the asset groups. The dynamic asset classification model is used to dynamically adjust the asset attributes of the asset data obtained in real time. The obtained asset attribute data includes:

[0100] Based on the key asset characteristics, the optimal number of clusters of the K-means clustering algorithm is determined by the elbow method, and unsupervised cluster analysis is performed on the key asset characteristics according to the optimal number of clusters to obtain the asset clustering results;

[0101] Generate asset group labels based on asset clustering results, train an extreme gradient boosting model based on the asset group labels, and use the trained extreme gradient boosting model to build a dynamic asset classification model;

[0102] The dynamic asset classification model is used to dynamically adjust the asset attributes of the asset data obtained in real time to obtain asset attribute data.

[0103] In this optional embodiment, based on the behavior assessment data and the asset attribute data, a multi-objective optimization algorithm is used to optimize asset allocation, and the generated optimized asset allocation plan includes:

[0104] Constructing several optimization objectives based on asset attribute data, and generating an optimization objective function using a linear weighted summation method based on the several optimization objectives;

[0105] Obtain user risk assessment results based on behavioral assessment data, determine the user's risk tolerance based on the user risk assessment results, and convert the user's risk tolerance into constraints for the multi-objective optimization algorithm;

[0106] The genetic algorithm is used to generate a set of Pareto optimal solutions in combination with an optimization objective function and constraint conditions, and the optimal solution distance method is used to optimize asset allocation based on the set of Pareto optimal solutions to generate an optimized asset allocation scheme.

[0107] In this alternative embodiment, the genetic algorithm is used to generate a set of Pareto optimal solutions in combination with an optimization objective function and constraint conditions, and the optimal solution distance method is used to optimize asset allocation based on the set of Pareto optimal solutions to generate an optimized asset allocation scheme, which includes:

[0108] The genetic algorithm is used to perform multi-objective optimization on the optimization objective function and constraint conditions, and a set of Pareto optimal solutions is generated based on the multi-objective optimization results;

[0109] The set of Pareto optimal solutions is standardized to obtain a set of standardized Pareto optimal solutions, and the optimal solution distance method is used to determine the positive ideal solution and the negative ideal solution based on the set of standardized Pareto optimal solutions;

[0110] The Euclidean distance of each Pareto optimal solution to the positive ideal solution and the negative ideal solution is calculated, and the proximity of the Pareto optimal solution to the positive ideal solution is calculated based on the calculated Euclidean distance;

[0111] The Pareto optimal solution with the highest proximity is selected for asset allocation optimization to generate an optimized asset allocation scheme.

[0112] In this alternative embodiment, a machine learning model is used to predict market trends based on pre-acquired market economic data, and a capital asset pricing model is used to extract market sensitivity in combination with the market trend prediction results, and the optimized asset allocation scheme is adjusted based on the market sensitivity to generate a final asset allocation scheme, which includes:

[0113] The pre-acquired market economic data is processed for missing values and outliers to obtain cleaned market economic data, and the cleaned market economic data is normalized and time-aligned to obtain standardized market economic data;

[0114] The standardized market economic data is divided into a training set and a test set, and the long short-term memory network model is trained using the training set, and the real-time acquired market economic data is used to predict market trends based on the trained long short-term memory network model to obtain market trend prediction results;

[0115] The capital asset pricing model is used to calculate the market sensitivity of each type of asset in combination with the market trend prediction results, the performance expectations of each type of asset are determined based on the market sensitivity of each type of asset, and the optimized asset allocation scheme is adjusted based on the performance expectations of each type of asset to generate a final asset allocation scheme.

[0116] In this optional embodiment, the capital asset pricing model is used in combination with market trend forecast results to calculate the market sensitivity of each asset class, and the performance expectations of each asset class are determined based on the market sensitivity of each asset class. Based on the performance expectations of each asset class, the optimized asset allocation plan is adjusted to generate a final asset allocation plan, including:

[0117] Based on the capital asset pricing model, a single-factor linear regression analysis is conducted on the market trend forecast results, and the market sensitivity of various assets is calculated using the single-factor linear regression analysis results;

[0118] Determine the performance expectations of each asset class based on its market sensitivity, identify priority assets based on the expected performance results of each asset class, and obtain priority assets;

[0119] Based on the priority assets, the asset ratio of the optimized asset allocation plan is dynamically adjusted to generate the final asset allocation plan.

[0120] Figure 2 An embodiment of the financial asset allocation system in the banking and insurance field based on intelligent analysis of the present invention is shown.

[0121] In this optional embodiment, the financial asset allocation system for the banking and insurance sector based on intelligent analysis includes: a behavior assessment data acquisition module 201, an asset attribute data acquisition module 202, an asset allocation plan generation module 203, and an asset allocation plan adjustment module 204;

[0122] The behavior evaluation data acquisition module 201 is used to acquire historical transaction data, pre-process the historical transaction data, and use the entropy method to combine the pre-processed historical transaction data to perform transaction behavior evaluation to obtain behavior evaluation data;

[0123] The asset attribute data acquisition module 202 is used to pre-process the pre-acquired asset data and dynamically adjust the asset attributes of the real-time acquired asset data using a dynamic asset classification model based on the pre-processed asset data to obtain asset attribute data;

[0124] The asset allocation plan generation module 203 is used to optimize asset allocation based on the behavior evaluation data and asset attribute data using a multi-objective optimization algorithm to generate an optimized asset allocation plan;

[0125] The asset allocation plan adjustment module 204 is used to use a machine learning model to predict market trends based on pre-acquired market economic data, and to extract market sensitivity using the capital asset pricing model combined with the market trend prediction results. Based on the market sensitivity, the optimized asset allocation plan is adjusted to generate a final asset allocation plan.

[0126] It should be noted that user information data includes user age, gender, age and occupation of the user's children, income level, total investment amount, etc.; historical transaction behavior data includes transaction product number, transaction amount, transaction years, transaction product type, transaction product income data, etc.; questionnaire data includes acceptable transaction amount, acceptable transaction years, transaction product preference type, etc.

[0127] It should be noted that the multi-dimensional behavioral characteristic indicators include average investment amount, investment frequency, average holding period, proportion of risky products, return volatility, risk preference, and liquidity preference.

[0128] It should be noted that for positive indicators (the bigger the better): Where x′ ij1 Represents the normalized value of the positive indicator; x ij1 Indicates the positive indicator value; max(x j ) represents the maximum index value; min(x j ) represents the minimum indicator value.

[0129] For negative indicators (smaller the better): Where x′ ij2 Indicates the normalized value of the negative indicator; x ij2 Indicates a negative indicator value; max(x j ) represents the maximum index value; min(x j ) represents the minimum indicator value.

[0130] It should be noted that the calculation formula for the proportion of each indicator is:

[0131]

[0132] Where p ij Indicates the proportion of indicators; m indicates the number of indicators.

[0133] It should be noted that the calculation formula for information entropy is:

[0134]

[0135] Where, e j represents the information entropy value; k represents a constant term.

[0136] Among them, when p ij = 0, define p ij ln(p ij )=0.

[0137] It should be noted that the calculation formula for the coefficient of difference is: g j =1-e j .

[0138] It should be noted that the weight calculation formula is: Where n represents the total number of coefficients of variation.

[0139] It should be noted that the extraction of a multi-dimensional customer portrait model (i.e., behavioral assessment data) involves analyzing historical transaction behaviors and calculating the risk preference coefficient using the entropy method through historical questionnaire surveys.

[0140] The specific code example is as follows:

[0141] class CustomerProfile:

[0142] def__init__(self,financial_data):

[0143] self.assets = {

[0144] 'current':financial_data['current'],

[0145] 'fixed':financial_data['regular'],

[0146] 'wealth_mgmt':financial_data['financial management'],

[0147] 'funds':financial_data['funds'],

[0148] 'insurance':financial_data['insurance'],

[0149] 'lump_sum':financial_data['lump sum']

[0150] }

[0151] self.total_assets=sum(self.assets.values())

[0152] self.risk_profile = self.calculate_risk_profile() # Based on questionnaire / trading behavior analysis

[0153] defcalculate_risk_profile(self):

[0154] #Use entropy method to calculate risk preference coefficient

[0155] volatility=[...]#Historical volatility of each asset

[0156] return np.dot(list(self.assets.values()),volatility) / self.total_assets.

[0157] It should be noted that the pre-acquired asset data includes asset type, asset price, asset transaction volume, asset return rate, asset volatility, etc.

[0158] It should be noted that asset attribute characteristics include average return, average trading volume, maximum drawdown, yield, volatility, turnover rate, etc.

[0159] It should be noted that a dynamic asset classification engine (i.e., a dynamic asset classification model) is constructed to dynamically adjust asset attribute parameters. Specifically, based on the clustering results, different attribute parameters can be assigned to each type of asset, such as risk level, liquidity score, expected return, etc.

[0160] It should be noted that the specific code example of the multi-objective optimization core algorithm is as follows:

[0161] defportfolio_optimization(customer,market_condition):

[0162] #Build optimization problem

[0163] model = ConcreteModel()

[0164] #Decision variables: allocation ratio of various assets

[0165] model.x=Var(asset_classes,bounds=(0,1))

[0166] #Objective function: three-objective weighted optimization

[0167] model.obj=Objective(expr=0.4*return_objective(model.x)

[0168] -0.3*risk_objective(model.x)+

[0169] 0.3*diversity_objective(model.x),sense=maximize)

[0170] #Constraints

[0171] model.total=Constraint(expr=sum(model.x.values())==1)

[0172] model.risk_limit=Constraint(expr=risk_objective(model.x)<=customer.risk_tolerance)

[0173] #Solve using genetic algorithm

[0174] solver=SolverFactory('genetic_algorithm')

[0175] results = solver.solve(model)

[0176] return optimized_allocation.

[0177] It should be noted that in order to calculate the market sensitivity of various assets based on market trend forecast results, it is necessary to introduce market indices or economic indicators (i.e., market economic data) to dynamically adjust the weights of different assets. The specific code example is as follows:

[0178] class MarketAdjuster:

[0179] def__init__(self):

[0180] self.factors = {

[0181] 'interest_rate':0.5,

[0182] 'cpi':0.3,

[0183] 'equity_risk_premium':0.7

[0184] }

[0185] defcalculate_weight_adjustment(self):

[0186] #Using LSTM to predict market factors

[0187] market_model=load_1stm_model()

[0188] predicted_factors=market_model.predict(next_quarter())

[0189] #Calculate dynamic smoothing coefficient

[0190] adjustment={

[0191] 'insurance':0.5*predicted_factors['interest_rate'],

[0192] 'funds':0.3*predicted_factors['equity_risk_premium'],

[0193] #Other asset adjustment logic...

[0194] }

[0195] return adjustment.

[0196] It's important to note that the visualization analysis engine needs to generate comparison charts before and after optimization, particularly radar charts, to show changes in metrics like profitability, security, and flexibility. This may require the support of a data visualization library, such as Matplotlib or D3.js. This part likely involves outputting data structures for front-end calls. The specific code example is as follows:

[0197] defgenerate_radar_chart(original,optimized):

[0198] categories = ['yield', 'risk control', 'liquidity', 'diversity', 'stability']

[0199] original_scores=[

[0200] calculate_return(original),

[0201] 1-calculate_risk(origina1),

[0202] calculate_liquidity(original),

[0203] calculate_diversity(original),

[0204] calculate_stability(original) ]

[0206] optimized_scores = [... ] # same as optimized scores

[0207] # generate radar chart using polar plot

[0208] fig = plt.figure(figsize=(8, 8))

[0209] ax = fig.add_subplot(111, polar=True)

[0210] ax.plot(theta, original_scores, color='red')

[0211] ax.plot(theta, optimized_scores, color='blue')

[0212] return fig.

[0213] It should be noted that the implementation steps in actual application: data collection and processing, multi-objective optimization, dynamic weight adjustment, comparative analysis and visualization, system implementation. At the same time, in the application, the model needs to be ensured to be interpretable, so that customers and customer managers can understand the basis of the optimization suggestions. At the same time, the model needs to be backtested by historical data to verify its effectiveness and stability, so as to avoid giving unreasonable suggestions.

[0214] It should be noted that the technical architecture includes data layer, algorithm layer and display layer; the data layer includes customer database and market data API; the algorithm layer includes optimization engine and market prediction model; the display layer includes Web service and visualization component.

[0215] Among them, the dynamic parameter adjustment mechanism is to update the asset attributes in real time combined with macroeconomic indicators; the third-order optimization target is the joint optimization of yield, risk or diversity; the customer-market two-dimensional analysis considers both individual characteristics and market changes.

[0216] In addition, the verification method is to use historical data backtesting to verify the effectiveness of the model, such as the initial assets of the example customer of 1 million, and the specific verification code example is as follows:

[0217] {

[0218] 'current': 40%, # 1.5% annualized for current account

[0219] 'fixed': 30%, # 3.2% annualized for fixed account

[0220] 'wealth_mgmt': 15%, #wealth management annualized rate 4.7%

[0221] 'funds': 5%, #funds expected annualized return 8%

[0222] 'insurance':8%,#insurance annualized rate 3.5%

[0223] 'lump_sum': 2% # Fixed annualized rate of 4.5%

[0224] }

[0225] #Market parameter adjustment example

[0226] market_index={

[0227] 'interest_rate':-0.5,#interest rate drops by 0.5%

[0228] 'equity_risk_premium': +2% #Stock market risk premium rises

[0229] }

[0230] #Dynamic polishing of asset attributes:

[0231] asset_properties['current']['return']-=0.5%#current deposit interest rate cut

[0232] asset_properties['funds']['return']+=1.2% #Expected return on equity assets increased

[0233] asset_properties['wealth_mgmt']['risk']-=0.1#Financial management risk coefficient decreases

[0234] #Original alliance expected income:

[0235] (40%*1.5%)+(30%*3.2%)+...=approximately RMB 28,300 annualized

[0236] #Suggested alliance after model optimization:

[0237] 'current': 25% (-15%) → Release liquidity

[0238] 'fixed': 25% (-5%) → Reduce low-yield fixed income

[0239] 'wealth_mgmt': 20% (+5%) → Increase allocation to high-quality financial products

[0240] 'funds': 15% (+10%) → Capture stock market opportunities

[0241] 'Insurance': 12% (+4%) → Lock in long-term gains

[0242] 'lump_sum': 3% (+1%) → Enhanced security

[0243] }

[0244] #Expected benefits after optimization:

[0245] (25%*1.0%)+(25%*3.2%)+(20%*4.5%)+

[0246] (15%*9.2%)+(12%*3.8%)+(3%*4.5%)=annualized approximately 36,700 yuan.

[0247] The verification result showed that the profit increased from 28,300 to 36,700.

[0248] The following is a code example for setting the risk threshold protection mechanism:

[0249] #Add risk constraints during optimization:

[0250] risk_constraint=sum([

[0251] asset_risk[class]*allocation[class]

[0252] forclassinasset_classes

[0253] ])<=customer.risk_tolerance*0.9#Retain a 10% safety margin

[0254] #Stress Test:

[0255] ifmarket_crisis_scenario:

[0256] funds_allocation = max(10%, current_allocation) #Set the alliance base position limit.

[0257] It needs to be explained that the application quantifies the indicators of each attribute, such as risk, liquidity, by calculating through industry standards or historical data. Multi-objective optimization uses methods such as genetic algorithms to find Pareto optimal solutions. The customer's risk assessment results are integrated into the model, and the customer's risk tolerance is converted into the constraints or target weights of the optimization model; finally, the part of the analysis report is output in a templated manner, combined with the customer's specific data and optimization results, to automatically generate suggestions. The entire system requires modular design to facilitate subsequent expansion and maintenance, and each module works independently but cooperatively.

[0258] As Figure 4 shown, the flow description of the user login system management flowchart is to enter the user login interface after starting the system, input the account password for login, verify whether it is a system user through logical judgment, if not, the user does not exist, then prompt the system error operation to terminate; if yes, the user exists, then allow to enter the operating system, and jump to the main management interface; the main system includes a user system and a management system; wherein the user system includes a data analysis and demonstration system; the management system includes a data rule maintenance management system, until the operation is executed.

[0259] As Figure 5 shown, the components and flow of the Torner engine user login verification logic diagram are that the user accesses the login interface at the user end, inputs the account password and submits it to the Tomcat server; the Tomcat server transfers the request to the login interface; the user clicks login after inputting the account password in the login interface; the UserController obtains the account password and calls the UserLogin method for processing; then, the UserService queries the database by calling the SelectUserByIdAndPassword method of the UserDAO to obtain user information.

[0260] Among them, UserController obtains the account and password entered by the user; UserController calls the UserLogin method of UserService to perform login processing; UserService calls the SelectUserByIdAndPassword method of UserDAO to query user information based on the account and password; UserDAO performs database query operations and calls the SQL statement in UserMapper.XML; UserDAO returns the User object based on the database query results; UserMapper.XML returns the queried User object to UserDAO; UserDAO returns the User object to UserService; after receiving the User object, UserService determines whether the User object is empty; if the User object is not empty, it returns True, jumps to the home page, and stores the UserId in the Session; if the User object is empty, it returns False, prompts an error message, and asks the user to return to the login page; UserMapper.XML returns the queried User object to UserDAO, and then UserDAO returns it to UserService. The UserService checks whether the User object is empty. If it is, it returns True, redirects to the homepage, and stores the UserId in the Session. If it is empty, it returns False, displays an error message, and returns the user to the login page. The verification condition is that if the User object is not empty, it returns True, redirects to the homepage, and stores the UserId in the Session; if the User object is empty, it returns False, indicating that the login failed. The database interaction is to execute the query and return the User object. The server completes the authentication by comparing the results.

[0261] like Figure 6 As shown, the core modules of the system access layer architecture diagram include the communication protocol layer, API gateway, message bus and storage layer; the communication protocol layer includes support for WebSocket, HTTP / HTTPS, and TCP protocols to ensure the compatibility of data transmission in multiple scenarios; the API gateway includes integrated server-side functional modules, and the integrated server-side functional modules include REST interface, IPC (inter-process communication), MQ (message queue), security services, log monitoring, etc., providing routing management, global service scheduling and terminal management capabilities; the message bus and storage layer include cache system, database and file storage. Figure 6The complete technology stack from the access layer to the storage layer is presented. At the access layer, users can access the system through Android mobile devices (phones and tablets), iOS mobile devices (phones and tablets), and PCs. The interface layer provides multiple communication protocols, including WebSocket, HTTP / HTTPS, and TCP, and these interfaces are centrally managed through an API gateway. The service layer is divided into two parts: service communication and service management. Service communication includes components such as REST, RPC, and MQ. The business service module covers functions such as user maintenance, smart upgrades, data measurement, scenario planning, data inversion, statistical reporting, and security services. Basic service components include membership services, data retention, and statistical mining. Service management involves service configuration, registration discovery, and log monitoring. The storage layer includes a message bus, file storage, caches (such as Redis and Memcached), and databases (such as MongoDB and MySQL), providing data support and storage services for the entire system. In addition, the cache system, including Redis and Memcached, supports fast read and write of high-frequency data. The database uses MySQL and a specialized database (such as mengpdb) for persistent storage of structured data. File storage is used for unstructured data management.

[0262] like Figure 7 As shown in the figure, the real-time data service and processing flow chart includes data collection, data transmission, core services, and output and interface; data collection is to collect user data (age, gender, children's information, etc.) and business data (cost, years) through the front-end page (such as UI page selection) or interface (RestfulAPI, APP interface); data transmission is encrypted and transmitted to the server using the HTTP protocol, and stored in the database after decryption; the core services of data application are profit calculation, scenario planning, product rule configuration, etc., which generate analysis results based on real-time data; the data storage module is to classify and store product data (value, quality, rules), user benefit data and system configuration information; output and interface are to display the analysis results through the system page, and provide standardized interfaces (such as APP interface) to the outside world to support third-party system integration. Figure 7The complete process from real-time data collection to data services is demonstrated. First, user data (such as age, gender, and age of children), payment simulation data (such as fees, years), and product data (such as product selection) are selected through the UI page and transmitted to the server via the HTTP protocol; at the same time, system data (including the product's underlying value library, the product's underlying revenue database, product rule configuration, etc.) directly enters the management system to maintain system storage. In data storage, the server is responsible for data reception, verification, decryption, and storage of personnel information, payment information, and product information, and integrates this data with system data (such as the value of products in different time periods, the efficiency of people and goods at different ages, and the rules of different products). Subsequently, through query analysis, the generated data services are applied to system pages (such as revenue calculation, value-added calculation, scenario planning) and other services (such as Restful API, APP interface, and other interfaces), thereby realizing the comprehensive management and utilization of data.

[0263] Figure 5 The Torner engine implements an efficient authentication mechanism and combines it with dynamic session management to improve system security. Figure 6 The API gateway and message bus design supports high-concurrency and scalable microservice architecture; Figure 7 The real-time data processing process ensures the accuracy of business analysis and data privacy through multi-dimensional data integration and encrypted transmission.

[0264] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 3 As shown. The computer device includes a processor, a memory and a network interface connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store static information and dynamic information data. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, the steps in the embodiment of the financial asset allocation method in the banking and insurance field based on intelligent analysis are implemented.

[0265] Those skilled in the art will understand that Figure 3 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present invention and does not constitute a limitation on the computer device to which the solution of the present invention is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0266] In addition, the present invention also provides a computer device, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the steps in the embodiment of the above-mentioned financial asset allocation method in the banking and insurance field based on intelligent analysis are implemented.

[0267] In addition, the present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps in the embodiment of the above-mentioned financial asset allocation method in the banking and insurance field based on intelligent analysis are implemented.

[0268] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiments of the method for allocating financial assets in the banking and insurance field based on intelligent analysis can be implemented by instructing related hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned embodiments of the method for allocating financial assets in the banking and insurance field based on intelligent analysis. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention may include at least non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0269] The present invention is not limited to the structures described above and shown in the drawings, and various modifications and changes can be made without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. A financial asset allocation method in the banking and insurance sector based on intelligent analysis, characterized by: include: Obtain historical transaction data, pre-process the historical transaction data, and use the entropy method to combine the pre-processed historical transaction data to evaluate transaction behavior and obtain behavior evaluation data; Preprocess the pre-acquired asset data, and dynamically adjust the asset attributes of the real-time acquired asset data using a dynamic asset classification model based on the pre-processed asset data to obtain asset attribute data; Based on behavioral assessment data and asset attribute data, we use multi-objective optimization algorithms to optimize asset allocation and generate optimized asset allocation plans; Based on the pre-acquired market economic data, the machine learning model is used to predict market trends, and the capital asset pricing model is used in combination with the market trend prediction results to extract market sensitivity. Based on the market sensitivity, the optimized asset allocation plan is adjusted to generate the final asset allocation plan.

2. The method for allocating financial assets in the banking and insurance sector based on intelligent analysis according to claim 1, characterized in that: The historical transaction data is obtained, pre-processed, and the transaction behavior evaluation is performed using the entropy method in combination with the pre-processed historical transaction data to obtain the behavior evaluation data including: Obtain historical transaction data and use interpolation to process missing values ​​in the historical transaction data to obtain completed historical transaction data; The historical transaction data includes user information data, historical transaction behavior data and questionnaire data; Perform outlier processing on the padded historical transaction data based on the interquartile range to obtain cleaned historical transaction data, and normalize the cleaned historical transaction data using the standard score to obtain pre-processed historical transaction data; Based on the pre-processed historical transaction data, statistical analysis methods are used to extract multi-dimensional behavioral characteristic indicators, and a behavioral evaluation indicator system is constructed using the multi-dimensional behavioral characteristic indicators; Based on the behavioral evaluation index system, the entropy method is used to calculate the weight of each indicator, and the multi-dimensional behavioral characteristic indicators are comprehensively evaluated using weighted summation according to the weight of each indicator to obtain behavioral evaluation data.

3. The method for allocating financial assets in the banking and insurance sector based on intelligent analysis according to claim 2 is characterized in that: Based on the behavior evaluation index system, the entropy method is used to calculate the weight of each indicator, and the multi-dimensional behavior characteristic indicators are comprehensively evaluated by weighted summation according to the weight of each indicator. The behavior evaluation data obtained includes: Based on the behavioral evaluation index system, the entropy method is used to calculate the proportion of each indicator, and the information entropy value of each indicator is calculated according to the proportion of each indicator; The difference coefficient of each indicator is calculated through the information entropy value of each indicator, and the weight of each indicator is obtained by using the difference coefficient of each indicator; According to the weight of each indicator, the multi-dimensional behavioral characteristic indicators are comprehensively evaluated using weighted summation to obtain behavioral evaluation data.

4. The method for allocating financial assets in the banking and insurance sector based on intelligent analysis according to claim 1, characterized in that: The asset attribute data obtained in advance is pre-processed, and the asset attributes of the real-time acquired asset data are dynamically adjusted using a dynamic asset classification model based on the pre-processed asset data. The obtained asset attribute data includes: Process missing values ​​and outliers on the pre-acquired asset data to obtain cleaned asset data, normalize the cleaned asset data using standard scores to obtain standard asset data, and perform time alignment on the standard asset data to obtain pre-processed asset data; Based on the pre-processed asset data, time series modeling is used to extract asset attribute characteristics, and the importance of asset attribute characteristics is evaluated using the Shapley additive interpretation method. Based on the importance evaluation results, the asset attribute characteristics are screened to obtain key asset characteristics; The K-means clustering algorithm is used to perform unsupervised cluster analysis on key asset features to obtain asset groups, and a dynamic asset classification model is constructed based on the asset groups. The dynamic asset classification model is used to dynamically adjust the asset attributes of the asset data obtained in real time to obtain asset attribute data.

5. The method for allocating financial assets in the banking and insurance sector based on intelligent analysis according to claim 4 is characterized in that: The K-means clustering algorithm is used to perform unsupervised cluster analysis on key asset features to obtain asset groups, and a dynamic asset classification model is constructed based on the asset groups. The dynamic asset classification model is used to dynamically adjust the asset attributes of the asset data obtained in real time, and the asset attribute data obtained includes: Based on the key asset characteristics, the optimal number of clusters of the K-means clustering algorithm is determined by the elbow method, and unsupervised cluster analysis is performed on the key asset characteristics according to the optimal number of clusters to obtain the asset clustering results; Generate asset group labels based on asset clustering results, train an extreme gradient boosting model based on the asset group labels, and use the trained extreme gradient boosting model to build a dynamic asset classification model; The dynamic asset classification model is used to dynamically adjust the asset attributes of the asset data obtained in real time to obtain asset attribute data.

6. The method for allocating financial assets in the banking and insurance sector based on intelligent analysis according to claim 1, characterized in that: The asset allocation optimization is performed based on the behavior assessment data and the asset attribute data using a multi-objective optimization algorithm to generate an optimized asset allocation plan including: Constructing several optimization objectives based on asset attribute data, and generating an optimization objective function using a linear weighted summation method based on the several optimization objectives; Obtain user risk assessment results based on behavioral assessment data, determine the user's risk tolerance based on the user risk assessment results, and convert the user's risk tolerance into constraints for the multi-objective optimization algorithm; Genetic algorithms are used in combination with optimization objective functions and constraints to generate a Pareto optimal solution set. Based on the Pareto optimal solution set, the superior and inferior solution distance method is used to optimize asset allocation and generate an optimized asset allocation plan.

7. The method for allocating financial assets in the banking and insurance sector based on intelligent analysis according to claim 6 is characterized in that: The method of using a genetic algorithm in combination with an optimization objective function and constraints to generate a Pareto optimal solution set, and optimizing asset allocation using a superior-inferior solution distance method based on the Pareto optimal solution set, generates an optimized asset allocation plan including: Use genetic algorithms to perform multi-objective optimization on the optimization objective function and constraints, and generate a Pareto optimal solution set based on the multi-objective optimization results; The Pareto optimal solution set is standardized to obtain a standardized Pareto optimal solution set, and the positive ideal solution and the negative ideal solution are determined based on the standardized Pareto optimal solution set using the superior and inferior solution distance method; Calculate the Euclidean distance of each Pareto optimal solution to the positive ideal solution and the negative ideal solution, and calculate the similarity between the Pareto optimal solution and the positive ideal solution based on the calculated Euclidean distance; Select the Pareto optimal solution with the highest similarity to optimize asset allocation and generate an optimized asset allocation plan.

8. The method for allocating financial assets in the banking and insurance sector based on intelligent analysis according to claim 1, characterized in that: The method of using a machine learning model to predict market trends based on pre-acquired market economic data, extracting market sensitivity using the capital asset pricing model combined with the market trend prediction results, and adjusting the optimized asset allocation plan based on the market sensitivity to generate the final asset allocation plan includes: Processing missing values ​​and outliers on pre-acquired market economic data to obtain cleaned market economic data, and normalizing and time-aligning the cleaned market economic data to obtain standardized market economic data; The standardized market economic data is divided into a training set and a test set, and the long short-term memory network model is trained using the training set. The market trend prediction is then performed on the real-time acquired market economic data based on the trained long short-term memory network model to obtain the market trend prediction results. The capital asset pricing model is used in combination with market trend forecast results to calculate the market sensitivity of various assets. The performance expectations of various assets are judged based on the market sensitivity of various assets. Based on the performance expectations of various assets, the optimized asset allocation plan is adjusted to generate the final asset allocation plan.

9. The method for allocating financial assets in the banking and insurance sector based on intelligent analysis according to claim 8, characterized in that: The aforementioned method utilizes the capital asset pricing model in combination with market trend forecast results to calculate the market sensitivity of various assets, determines the performance expectations of various assets based on the market sensitivity of various assets, and adjusts the optimized asset allocation plan based on the performance expectations of various assets to generate the final asset allocation plan, including: Based on the capital asset pricing model, a single-factor linear regression analysis is conducted on the market trend forecast results, and the market sensitivity of various assets is calculated using the single-factor linear regression analysis results; Determine the performance expectations of each asset class based on its market sensitivity, identify priority assets based on the expected performance results of each asset class, and obtain priority assets; Based on the priority assets, the asset ratio of the optimized asset allocation plan is dynamically adjusted to generate the final asset allocation plan.

10. A financial asset allocation system in the banking and insurance field based on intelligent analysis, used to implement the financial asset allocation method in the banking and insurance field based on intelligent analysis as described in any one of claims 1 to 9, characterized in that: The financial asset allocation system in the banking and insurance field based on intelligent analysis includes: a behavior assessment data acquisition module, an asset attribute data acquisition module, an asset allocation plan generation module, and an asset allocation plan adjustment module; The behavior evaluation data acquisition module is used to acquire historical transaction data, pre-process the historical transaction data, and use the entropy method to combine the pre-processed historical transaction data to perform transaction behavior evaluation to obtain behavior evaluation data; The asset attribute data acquisition module is used to pre-process the pre-acquired asset data and dynamically adjust the asset attributes of the real-time acquired asset data using a dynamic asset classification model based on the pre-processed asset data to obtain asset attribute data; The asset allocation plan generation module is used to optimize asset allocation based on the behavior evaluation data and asset attribute data using a multi-objective optimization algorithm to generate an optimized asset allocation plan; The asset allocation plan adjustment module is used to use a machine learning model to predict market trends based on pre-acquired market economic data, and to extract market sensitivity using a capital asset pricing model combined with market trend prediction results, and to adjust the optimized asset allocation plan based on market sensitivity to generate a final asset allocation plan.

Images