Financial data self-adaptive optimization processing method and system based on AI agent

Abstract

The application discloses a financial data self-adaptive optimization processing method and system based on an AI agent, and belongs to the technical field of optimization processing; all financial indicators in a financial service platform are called, and user interaction behavior track data of the financial indicators is acquired; a financial explicit data set and a financial implicit data set are constructed, a cognitive bias factor of a user to a single financial indicator at a single time point is calculated, and a cognitive balance weight and a cognitive response intensity value of the user to the single financial indicator at the single time point are calculated; based on the cognitive response intensity value, a time point overall cognitive bias and a financial indicator overall cognitive bias are calculated; based on the cognitive response intensity value, the time point overall cognitive bias and the financial indicator overall cognitive bias, a bias fusion intensity coefficient of the user is calculated; based on the bias fusion intensity coefficient, analysis and early warning are performed, so that the comprehensive effects of improving user financial behavior recognition accuracy, enhancing platform risk control capability and optimizing personalized service configuration are achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of optimization processing, in particular to a financial data adaptive optimization processing method and system based on an AI agent. BACKGROUND

[0002] In recent years, AI agents have been introduced into the financial data processing system, not only improving the automation level of data analysis, but also enhancing the adaptive ability of financial services. At the same time, with the significant increase in the number of financial users and interaction frequency, the types of data obtained by the platform have expanded from traditional structured market data to unstructured information such as user behavior trajectories, psychological preferences, and interaction feedback. Financial data presents complex characteristics of high dimensionality, dynamics, and heterogeneity coexistence. In order to extract valuable information from multi-source, multi-dimensional financial data, researchers have attempted to introduce methods based on reinforcement learning, self-attention mechanisms, and deep behavior modeling to build more detailed user portraits and risk judgment models. However, most existing methods focus on modeling explicit data (such as prices, trading volumes, and financial indicators), and the understanding and modeling of cognitive biases at the user level are still weak, making it difficult to fully depict the interaction mechanism between users and the market.

[0003] There are several deficiencies in the existing technology in modeling user cognitive behavior. First, the current mainstream methods generally ignore the cognitive bias structure between user behavior and market indicators, do not effectively quantify the decision deviation of users due to psychological factors or experience limitations, and thus have blind spots in risk identification at the user level. Second, existing methods lack alignment mechanisms with market explicit data when processing user behavior data, and cannot dynamically explain the interaction behavior under the same indicator at the same time point. In addition, most existing analysis processes are static analysis, lacking bias tracking and early warning capabilities in the time dimension, and cannot reflect the changing trend of user cognitive state in real time. Therefore, there are still problems such as low precision and delayed response in key areas such as user understanding, behavior judgment, and risk prompt. SUMMARY

[0004] The present application aims to provide a financial data adaptive optimization processing method and system based on an AI agent to solve the problems raised in the background technology.

[0005] To solve the above technical problems, the present application provides the following technical solutions:

[0006] The application discloses an AI agent-based financial data adaptive optimization processing method, which comprises the following steps: S1, acquiring all financial indicators in a financial service platform and user interaction behavior trajectory data of the financial indicators; constructing a financial explicit data set and a financial implicit data set; S2, calculating a cognitive bias factor of a user at a single time point for a single financial indicator based on the financial explicit data set and the financial implicit data set; S3, calculating a cognitive balance weight and a cognitive response intensity value of the user at the single time point for the single financial indicator based on the cognitive bias factor; calculating a time point overall cognitive bias and a financial indicator overall cognitive bias based on the cognitive response intensity value; S4, calculating a bias fusion intensity coefficient of the user based on the cognitive response intensity value, the time point overall cognitive bias and the financial indicator overall cognitive bias; and performing analysis and early warning based on the bias fusion intensity coefficient.

[0007] As a preferred scheme of the AI agent-based financial data adaptive optimization processing method, all financial indicators in a financial service platform are acquired by an AI agent, static content data of the financial indicators is acquired and normalized, the financial indicators include market quotation data, asset price data, product data and risk indicator data, the static content data includes historical transaction data, financial indicator data and economic data of the financial indicators, the static content data of the financial indicators is time point aligned and marked as a financial explicit data set.

[0008] After user authorization, user interaction behavior trajectory data of the financial indicators is acquired from a financial service platform backend by an AI agent and normalized, the interaction behavior trajectory data includes access frequency data, transaction operation data and stay time data, the interaction behavior trajectory data is time point aligned and marked as a financial implicit data set.

[0009] As a preferred scheme of the AI agent-based financial data adaptive optimization processing method, the financial explicit data set and the financial implicit data set of an a-th financial indicator corresponding to an i-th time point are respectively denoted as {HT i (SC a ),FI i (SC a ),ED i (SC a )} and {VD i (SC a ),TO i (SC a ),DT i (SC a )}, wherein HT i (SC a) represents the historical transaction data of the a-th financial indicator corresponding to the i-th time point, FI i (SC a ) represents the financial indicator data of the a-th financial indicator corresponding to the i-th time point, ED i (SC a ) represents the economic data of the a-th financial indicator corresponding to the i-th time point, VD i (SC a ) represents the access frequency data of the a-th financial indicator corresponding to the i-th time point, TO i (SC a ) represents the transaction operation data of the a-th financial indicator corresponding to the i-th time point, DT i (SC a ) represents the dwell time data of the a-th financial indicator corresponding to the i-th time point, SC a represents the a-th financial indicator;

[0010] Based on the financial explicit data set {HT i (SC a ), FI i (SC a ), ED i (SC a )} and the financial implicit data set {VD i (SC a ), TO i (SC a ), DT i (SC a )}, the cognitive bias factor of the user at the i-th time point for the financial indicator SC a is calculated, and the calculation formula is as follows:

[0011]

[0012] wherein CBF i (SC a ) represents the cognitive bias factor of the user at the i-th time point for the financial indicator SC a , and ∈ represents a preset constant coefficient;

[0013] It should be noted that the cognitive bias factor is used to quantify the degree of deviation between user behavior (implicit data) and objective market reality (explicit data). Such deviation may be due to psychological factors of the user, such as overconfidence, attention deviation or decision deviation (for example, the user trades excessively or pays excessive attention when market volatility is low). The calculation of CBF i (SC a ) should combine implicit data and explicit data to generate a scalar value representing the specific time point ii and the specific financial indicator SC athe bias intensity.

[0014] As a preferred scheme of the AI agent-based financial data adaptive optimization processing method, the cognitive bias factor CBF a (SC i ) of the user at the i-th time point is calculated based on the cognitive bias of the user at the i-th time point to the financial indicator SC a . a The calculation formula is as follows:

[0015]

[0016] wherein CEW i (SC a ) represents the cognitive balance weight of the user at the i-th time point to the financial indicator SC a , and k represents a preset sensitivity coefficient.

[0017] Based on the cognitive balance weight CEW a (SC i ) of the user at the i-th time point to the financial indicator SC a and the cognitive bias factor CBF a (SC i ) of the user at the i-th time point to the financial indicator SC a , the cognitive response intensity value of the user at the i-th time point to the financial indicator SC a is calculated, and the calculation formula is: CRS i (SC a ) = CEW i (SC a ) × CBF i (SC a ), wherein CRS i (SC a ) represents the cognitive response intensity value of the user at the i-th time point to the financial indicator SC a .

[0018] The cognitive response intensity value of the user at the i-th time point to all financial indicators and the cognitive response intensity value of the user at all time points to the financial indicator SC a are obtained respectively, and the time point-based cognitive response intensity average and the financial indicator-based cognitive response intensity average are calculated, which are denoted as time point overall cognitive bias and financial indicator overall cognitive bias

[0019] As a preferred scheme of the AI agent-based financial data adaptive optimization processing method, the cognitive bias factor CBF acognitive reaction strength value CRS i (SC a ), point of time overall cognitive bias and financial indicator overall cognitive bias The deviation fusion strength coefficient of the user is calculated, and the calculation formula is as follows:

[0020]

[0021] Wherein, DFSC i (SC a ) represents the deviation fusion strength coefficient of the user at the i th time point to the financial indicator SC a .

[0022] If the deviation fusion strength coefficient DFSC a (SC i ) of the user at the i th time point to the financial indicator SC a is greater than 1 or less than 1, it is determined that the user has positive deviation or negative deviation at the i th time point to the financial indicator SC a , that is, overreaction or underreaction, and the relevant staff is warned;

[0023] All time points and all financial indicators are traversed for dynamic analysis and warning.

[0024] It should be noted that DFSC i (SC a ) greater than 1 indicates that the cognitive deviation strength of the user at the i th time point to the financial indicator SC a is higher than the expected benchmark, and the user may amplify the risk due to excessive attention (high VD i (SC a ) and DT i (SC a )) or irrational trading (high TO i (SC a )); DFSC i (SC a ) less than 1 indicates that the cognitive deviation strength of the user at the i th time point to the financial indicator SC a is lower than the expected benchmark, and the user may underestimate important indicators (such as ignoring financial report risks) or miss signals due to fatigue decision-making.

[0025] The financial data self-adaptive optimization processing system based on AI agent comprises a data acquisition and set construction module, a cognitive bias factor calculation module, a strength value and overall cognitive bias calculation module, and a strength coefficient calculation and analysis warning module.

[0026] The data acquisition and set construction module: acquires all financial indicators and user interaction behavior trajectory data of the financial indicators in the financial service platform; constructs a financial explicit data set and a financial implicit data set;

[0027] The cognitive bias factor calculation module: based on the financial explicit data set and the financial implicit data set, calculates the cognitive bias factor of the user at a single time point for a single financial indicator;

[0028] The intensity value and overall cognitive bias calculation module: based on the cognitive bias factor, calculates the cognitive balance weight and the cognitive response intensity value of the user at a single time point for a single financial indicator; based on the cognitive response intensity value, calculates the time point overall cognitive bias and the financial indicator overall cognitive bias;

[0029] The intensity coefficient calculation and analysis early warning module: based on the cognitive response intensity value, the time point overall cognitive bias and the financial indicator overall cognitive bias, calculates the bias fusion intensity coefficient of the user; based on the bias fusion intensity coefficient, performs analysis and early warning.

[0030] Further, the data acquisition and set construction module includes a data acquisition and set construction unit;

[0031] The data acquisition and set construction unit: based on an AI agent, acquires all financial indicators from the financial service platform, acquires static content data of the financial indicators and performs normalization processing, the financial indicators include market quotation data, asset price data, product data and risk indicator data, the static content data includes historical transaction data, financial indicator data and economic data of the financial indicators; time point aligns the static content data of the financial indicators, and marks as a financial explicit data set; after user authorization, uses an AI agent to acquire user interaction behavior trajectory data of the financial indicators from the financial service platform backend and performs normalization processing, the interaction behavior trajectory data includes access frequency data, transaction operation data and dwell time data; time point aligns the interaction behavior trajectory data, and marks as a financial implicit data set.

[0032] Further, the cognitive bias factor calculation module includes a cognitive bias factor calculation unit;

[0033] The cognitive bias factor calculation unit: based on the financial explicit data set and the financial implicit data set, calculates the cognitive bias factor of the user at the i-th time point for the a-th financial indicator.

[0034] Further, the intensity value and overall cognitive bias calculation module includes an intensity value calculation unit and an overall cognitive bias calculation unit;

[0035] The intensity value calculation unit: based on the cognitive bias factor of the user on the a-th financial indicator at the i-th time point, calculates the cognitive balanced weight of the user on the a-th financial indicator at the i-th time point; based on the cognitive balanced weight of the user on the a-th financial indicator at the i-th time point and the cognitive bias factor of the user on the a-th financial indicator at the i-th time point, calculates the cognitive response intensity value of the user on the a-th financial indicator at the i-th time point;

[0036] The overall cognitive bias calculation unit: respectively acquires the cognitive response intensity value of the user on all financial indicators at the i-th time point and the cognitive response intensity value of the user on the a-th financial indicator at all time points, and calculates the time point-based cognitive response intensity mean value and the financial indicator-based cognitive response intensity mean value, respectively denoted as time point overall cognitive bias and financial indicator overall cognitive bias.

[0037] Further, the intensity coefficient calculation and analysis early warning module comprises an intensity coefficient calculation unit and an analysis early warning unit;

[0038] The intensity coefficient calculation unit: based on the cognitive response intensity value of the user on the a-th financial indicator at the i-th time point, the time point overall cognitive bias and the financial indicator overall cognitive bias, calculates the bias fusion intensity coefficient of the user;

[0039] The analysis early warning unit: if the bias fusion intensity coefficient of the user on the a-th financial indicator at the i-th time point is greater than 1 or less than 1, it is determined that the user has positive bias or negative bias on the a-th financial indicator at the i-th time point, that is, overreaction or insufficient reaction, and then the related staff are warned; all time points and all financial indicators are traversed for dynamic analysis and early warning.

[0040] Compared with the prior art, the beneficial effects achieved by the present application are: in the AI agent-based financial data adaptive optimization processing method and system provided by the present application, comprehensive financial indicators and user interaction behavior trajectory data are obtained from a financial service platform, explicit data set and implicit data set are constructed, double-structured modeling of market objective data and user behavior data is realized, and basic guarantee is provided for subsequent cognitive analysis; the explicit and implicit data are used to quantify the cognitive bias of a user to a certain financial indicator at a specific time point, the psychological deviation between user behavior and objective market state is revealed, and a measurable index is provided for understanding the user investment behavior characteristics; by constructing a cognitive balance weight and cognitive response intensity model, dynamic decomposition and multi-dimensional analysis of user cognitive bias are realized, and overall cognitive bias in the time dimension and the index dimension is further obtained, and the overall behavior trend of the user and the cognitive tendency of the user on different financial indicators are accurately described; the bias fusion intensity coefficient is introduced, the individual response intensity and group bias characteristics are integrated, it is determined whether the user has positive or negative bias behavior, and an intelligent early warning strategy based on the AI agent is formed accordingly, and dynamic intervention and pre-response to the user's irrational decision risk are realized. Overall, the method realizes the comprehensive beneficial effects of improving the accuracy of user financial behavior identification, enhancing the platform risk control ability, and optimizing the configuration of personalized services through the organic combination of intelligent modeling, cognitive bias evaluation and fusion warning. BRIEF DESCRIPTION OF DRAWINGS

[0041] The accompanying drawings are included to provide a further understanding of the present application, and constitute a part of the specification, illustrate the present application together with the embodiments thereof, and explain the principles of the present application, and do not constitute a limitation of the present application.

[0042] Figure 1 is a step schematic diagram of the AI agent-based financial data adaptive optimization processing method of the present application;

[0043] Figure 2 is a structural schematic diagram of the AI agent-based financial data adaptive optimization processing system of the present application. DETAILED DESCRIPTION

[0044] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0045] Please refer to Figure 1 In the first embodiment, an AI agent-based financial data adaptive optimization processing method is provided, which comprises the following steps:

[0046] Step S1: Obtain all financial indicators and user interaction behavior trajectory data of the financial indicators from the financial service platform; construct a financial explicit dataset and a financial implicit dataset.

[0047] Specifically, based on the AI agent, all financial indicators are called from the financial service platform, the static content data of the financial indicators is obtained and normalized, the financial indicators include market quote data, asset price data, product data and risk indicator data, the static content data includes historical transaction data, financial indicator data and economic data of the financial indicators; the static content data of the financial indicators is time-point aligned and marked as a financial explicit dataset.

[0048] Further, after user authorization, the AI agent is used to call the user interaction behavior trajectory data of the financial indicators from the financial service platform backend and perform normalization processing, the interaction behavior trajectory data includes access frequency data, transaction operation data and dwell time data; the interaction behavior trajectory data is time-point aligned and marked as a financial implicit dataset.

[0049] Step S2: Based on the financial explicit dataset and the financial implicit dataset, the cognitive bias factor of the user to a single financial indicator at a single time point is calculated.

[0050] Specifically, the financial explicit dataset and the financial implicit dataset of the a-th financial indicator corresponding to the i-th time point are denoted as {HT i (SC a ),FI i (SC a ),ED i (SC a )} and {VD i (SC a ),TO i (SC a ),DT i (SC a )}, wherein HT i (SC a ) represents the historical transaction data of the a-th financial indicator corresponding to the i-th time point, FI i (SC a ) represents the financial indicator data of the a-th financial indicator corresponding to the i-th time point, ED i (SC a ) represents the economic data of the a-th financial indicator corresponding to the i-th time point, VD i (SC a ) represents the access frequency data of the a-th financial indicator corresponding to the i-th time point, TO i(SC a ) represents the transaction operation data of the a-th financial indicator corresponding to the i-th time point, DT i (SC a ) represents the dwell time data of the a-th financial indicator corresponding to the i-th time point, SC a represents the a-th financial indicator;

[0051] Further, based on the financial explicit data set {HT i (SC a ), FI i (SC a ), ED i (SC a )} and the financial implicit data set {VD i (SC a ), TO i (SC a ), DT i (SC a )}, the cognitive bias factor of the user at the i-th time point for the financial indicator SC a is calculated, and the calculation formula is as follows:

[0052]

[0053] Wherein, CBF i (SC a ) represents the cognitive bias factor of the user at the i-th time point for the financial indicator SC a , and ∈ represents a preset constant coefficient;

[0054] It should be noted that the cognitive bias factor is used to quantify the degree of deviation between the user behavior (implicit data) and the objective market reality (explicit data). Such deviation may be due to psychological factors of the user, such as overconfidence, attention deviation or decision deviation (for example, the user trades excessively or pays excessive attention when the market fluctuation is low). The calculation of CBF i (SC a ) should combine implicit data and explicit data to generate a scalar value representing the deviation intensity at a specific time point ii and a specific financial indicator SC a .

[0055] Step S3: based on the cognitive bias factor, the cognitive equilibrium weight and the cognitive response intensity value of the user at a single time point for a single financial indicator are calculated; based on the cognitive response intensity value, the overall cognitive bias at the time point and the overall cognitive bias for the financial indicator are calculated.

[0056] Specifically, based on the cognitive bias factor CBF a (SC i ) of the user at the i-th time point for the financial indicator SCa ), the cognitive balanced weight of the user at the i-th time point to the financial indicator SC a is calculated, and the calculation formula is as follows:

[0057]

[0058] wherein, CEW i (SC a ) represents the cognitive balanced weight of the user at the i-th time point to the financial indicator SC a , and k represents a preset sensitivity coefficient;

[0059] Further, based on the cognitive balanced weight CEW a (SC i ) of the user at the i-th time point to the financial indicator SC a and the cognitive bias factor CBF a (SC i ) of the user at the i-th time point to the financial indicator SC a , the cognitive response intensity value of the user at the i-th time point to the financial indicator SC a is calculated, and the calculation formula is: CRS i (SC a ) = CEW i (SC a ) × CBF i (SC a ), wherein CRS i (SC a ) represents the cognitive response intensity value of the user at the i-th time point to the financial indicator SC a ;

[0060] Further, the cognitive response intensity value of the user at the i-th time point to all financial indicators and the cognitive response intensity value of the user at all time points to the financial indicator SC a are respectively obtained, and the time point overall cognitive bias and the financial indicator overall cognitive bias

[0061] Step S4: based on the cognitive response intensity value, the time point overall cognitive bias and the financial indicator overall cognitive bias, the bias fusion intensity coefficient of the user is calculated; and based on the bias fusion intensity coefficient, analysis and early warning are performed.

[0062] Specifically, based on the cognitive response intensity value CRS i (SC a ) of the user at the i-th time point to the financial indicator SC a , the time point overall cognitive bias and financial indicator overall cognitive bias The bias fusion strength coefficient of the user is calculated, and the calculation formula is as follows:

[0063]

[0064] Wherein, DFSC i (SC a ) represents the bias fusion strength coefficient of the user to the financial indicator SC a at the i th time point;

[0065] Further, if the bias fusion strength coefficient DFSC a (SC i ) of the user to the financial indicator SC a at the i th time point is greater than 1 or less than 1, it is determined that the user has positive bias or negative bias, that is, overreaction or underreaction, to the financial indicator SC a at the i th time point, and relevant staff are warned;

[0066] In the present application, the formula combines three inputs: CRS i (SC a ) (cognitive bias strength of a specific time point and financial indicator), overall bias level at the i th time point (average of all financial indicators), and overall bias level of the financial indicator SC a (average of all time points); by dividing CRS i (SC a ) by the geometric mean of and , the formula quantifies the bias strength of a specific value relative to the overall background. If DFSC i (SC a ) is greater than 1, it indicates that the cognitive bias strength of the user to the financial indicator SC a at the i th time point is higher than the overall background (there may be significant psychological bias, such as overconfidence or attention bias), if DFSC i (SC a ) is equal to 1, it indicates that the bias strength is consistent with the overall background, and if DFSC i (SC a ) is less than 1, it indicates that the bias strength is lower than the overall background. This coefficient can be used for subsequent adaptive optimization, such as adjusting financial data weights or generating personalized recommendations, to reduce the impact of cognitive bias. The formula defines the benchmark as 1, because the denominator is essentially the expected bias benchmark, Fusion of "current overall state" and "index historical normal", constitute a dynamic baseline. Threshold 1, namely the current deviation intensity CRS i (SC a ) is exactly equal to the baseline.

[0067] The whole time point and all financial indicators are traversed, and the dynamic analysis warning is carried out.

[0068] Please refer to Figure 2 In the second embodiment: provide an AI agent-based financial data adaptive optimization processing system, which includes: data acquisition and set construction module, cognitive bias factor calculation module, intensity value and overall cognitive bias calculation module, and intensity coefficient calculation and analysis warning module;

[0069] The data acquisition and set construction module: acquires all financial indicators and user interaction behavior trajectory data of the financial indicators in the financial service platform; constructs financial explicit data set and financial implicit data set;

[0070] The cognitive bias factor calculation module: based on the financial explicit data set and the financial implicit data set, calculates the cognitive bias factor of the user to the single financial indicator at the single time point;

[0071] The intensity value and overall cognitive bias calculation module: based on the cognitive bias factor, calculates the cognitive balance weight and cognitive response intensity value of the user to the single financial indicator at the single time point; based on the cognitive response intensity value, calculates the time point overall cognitive bias and the financial indicator overall cognitive bias;

[0072] The intensity coefficient calculation and analysis warning module: based on the cognitive response intensity value, the time point overall cognitive bias and the financial indicator overall cognitive bias, calculates the bias fusion intensity coefficient of the user; based on the bias fusion intensity coefficient, carries out analysis and warning.

[0073] Further, the data acquisition and set construction module includes data acquisition and set construction unit;

[0074] The data acquisition and set construction unit: based on the AI agent, calls all financial indicators from the financial service platform, acquires static content data of the financial indicators and performs normalization processing, the financial indicators including market quotation data, asset price data, product data and risk indicator data, the static content data including historical transaction data, financial indicator data and economic data of the financial indicators; aligns the static content data of the financial indicators by time point, and labels as a financial explicit data set; after user authorization, uses the AI agent to call user interaction behavior trajectory data of the financial indicators from the financial service platform backend and performs normalization processing, the interaction behavior trajectory data including access frequency data, transaction operation data and dwell time data; aligns the interaction behavior trajectory data by time point, and labels as a financial implicit data set.

[0075] Further, the cognitive bias factor calculation module includes a cognitive bias factor calculation unit;

[0076] The cognitive bias factor calculation unit: based on the financial explicit data set and the financial implicit data set, calculates the cognitive bias factor of the user to the a-th financial indicator at the i-th time point.

[0077] Further, the intensity value and overall cognitive bias calculation module includes an intensity value calculation unit and an overall cognitive bias calculation unit;

[0078] The intensity value calculation unit: based on the cognitive bias factor of the user to the a-th financial indicator at the i-th time point, calculates the cognitive balance weight of the user to the a-th financial indicator at the i-th time point; based on the cognitive balance weight of the user to the a-th financial indicator at the i-th time point and the cognitive bias factor of the user to the a-th financial indicator at the i-th time point, calculates the cognitive response intensity value of the user to the a-th financial indicator at the i-th time point;

[0079] The overall cognitive bias calculation unit: respectively acquires the cognitive response intensity value of the user to all financial indicators at the i-th time point and the cognitive response intensity value of the user to the a-th financial indicator at all time points, and calculates the time point-based cognitive response intensity mean value and the financial indicator-based cognitive response intensity mean value, respectively labeled as time point overall cognitive bias and financial indicator overall cognitive bias.

[0080] Further, the intensity coefficient calculation and analysis warning module includes an intensity coefficient calculation unit and an analysis warning unit;

[0081] The intensity coefficient calculation unit: based on the cognitive response intensity value of the user to the a-th financial indicator at the i-th time point, the time point overall cognitive bias and the financial indicator overall cognitive bias, calculates the bias fusion intensity coefficient of the user;

[0082] The analysis early warning unit: if the deviation fusion intensity coefficient of the user to the a-th financial indicator at the i-th time point is greater than 1 or less than 1, it is determined that the user has positive deviation or negative deviation, that is, overreaction or insufficient reaction to the a-th financial indicator at the i-th time point, and early warning is given to the relevant staff; all time points and all financial indicators are traversed to perform dynamic analysis and early warning.

[0083] It should be noted that the relational terms herein such as first and second and the like are used solely to distinguish one entity or action from another, without necessarily requiring or implying any such actual relationship or order between such entities or actions. Moreover, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus.

[0084] Finally, it should be noted that: the above only describes the preferred embodiments of the present application, and is not intended to limit the present application, although the foregoing embodiments of the present application are described in detail, those skilled in the art can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features. Any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. An AI agent-based financial data adaptive optimization processing method, characterized in that, The method comprises the following steps: Step S1: acquiring all financial indicators and user interaction behavior trajectory data of the financial indicators from a financial service platform; constructing a financial explicit data set and a financial implicit data set; Step S2: calculating a cognitive bias factor of a user at a single time point for a single financial indicator based on the financial explicit data set and the financial implicit data set; Step S3: calculating a cognitive balance weight and a cognitive response intensity value of the user at the single time point for the single financial indicator based on the cognitive bias factor; and calculating a time point overall cognitive bias and a financial indicator overall cognitive bias based on the cognitive response intensity value; Step S4: calculating a bias fusion intensity coefficient of the user based on the cognitive response intensity value, the time point overall cognitive bias and the financial indicator overall cognitive bias; and performing analysis and early warning based on the bias fusion intensity coefficient. The specific implementation process of the step S2 comprises: Let the financial explicit data set and the financial implicit data set of the ath financial indicator corresponding to the ith time point be denoted as and wherein, represents the historical transaction data of the ath financial indicator corresponding to the ith time point, represents the financial indicator data of the ath financial indicator corresponding to the ith time point, represents the economic data of the ath financial indicator corresponding to the ith time point, represents the access frequency data of the ath financial indicator corresponding to the ith time point, represents the transaction operation data of the ath financial indicator corresponding to the ith time point, represents the dwell time data of the ath financial indicator corresponding to the ith time point, represents the ath financial indicator; Based on a financial explicit dataset and a financial implicit dataset , the cognitive bias factor of the user at the i-th time point for the financial indicator is calculated according to the following formula: ； wherein, represents a cognitive bias factor of the user at the i-th time point to the financial indicator , and represents a preset constant coefficient; The specific implementation process of the step S3 comprises: Based on the user's financial indicators at time point i Cognitive bias factor Calculate the user's response to financial indicators at time point i. The cognitive equilibrium weight is calculated using the following formula: ； wherein, represents the cognitive balance weight of the user at the i-th time point for the financial indicator , k represents a preset sensitivity coefficient; Based on the user's financial indicators at time point i Cognitive balance weight and the user's financial indicators at time point i Cognitive bias factor Calculate the user's response to financial indicators at time point i. The cognitive response intensity value is calculated using the following formula: ,in, This indicates that the user's view on financial indicators at time point i. The intensity value of cognitive response; Obtain the intensity of the user's cognitive response to all financial indicators at time point i and the intensity of the user's cognitive response to financial indicators at all time points. The cognitive response intensity values ​​were calculated, and the mean cognitive response intensity based on time points and the mean cognitive response intensity based on financial indicators were calculated, which were denoted as the overall cognitive bias at time points, respectively. Overall cognitive biases with financial indicators ; The specific implementation process of the step S4 comprises: Based on the user's financial indicators at time point i Cognitive response intensity value Overall cognitive bias at a given time point Overall cognitive biases with financial indicators The user's deviation fusion strength coefficient is calculated using the following formula: ； wherein, represents the deviation fusion intensity coefficient of the user at the i-th time point to the financial index ; If the user sets financial indicators at time point i... Deviation fusion strength coefficient If the value is greater than 1 or less than 1, then it is determined that the user's financial indicator at time point i is correct. If there is a positive or negative bias, i.e. overreaction or underreaction, a warning will be issued to the relevant staff. All time points and all financial indicators are traversed to perform dynamic analysis and early warning.

2. The AI agent-based financial data self-adaptive optimization processing method according to claim 1, characterized in that, The specific implementation process of the step S1 comprises: Based on an AI agent, all financial indicators are acquired from the financial service platform, static content data of the financial indicators is acquired and normalized, the financial indicators comprise market quotation data, asset price data, product data and risk indicator data, the static content data comprises historical transaction data, financial indicator data and economic data of the financial indicators; and the static content data of the financial indicators is time point aligned and marked as a financial explicit data set; After authorization by the user, the user interaction behavior trajectory data of the financial indicators is acquired from the financial service platform backend by the AI agent and normalized, the interaction behavior trajectory data comprises access frequency data, transaction operation data and dwell time data; and the interaction behavior trajectory data is time point aligned and marked as a financial implicit data set.

3. A financial data adaptive optimization processing system based on AI intelligent agents, executing the financial data adaptive optimization processing method based on AI intelligent agents as described in any one of claims 1-2, characterized in that, The system comprises a data acquisition and set construction module, a cognitive bias factor calculation module, an intensity value and overall cognitive bias calculation module, and an intensity coefficient calculation and analysis early warning module; The data acquisition and set construction module acquires all financial indicators and user interaction behavior trajectory data of the financial indicators from a financial service platform; and constructs a financial explicit data set and a financial implicit data set; The cognitive bias factor calculation module calculates a cognitive bias factor of a user at a single time point for a single financial indicator based on the financial explicit data set and the financial implicit data set; The intensity value and overall cognitive bias calculation module calculates a cognitive balance weight and a cognitive response intensity value of the user at a single time point for a single financial indicator based on the cognitive bias factor; and calculates a time point overall cognitive bias and a financial indicator overall cognitive bias based on the cognitive response intensity value; The intensity coefficient calculation and analysis early warning module calculates a bias fusion intensity coefficient of the user based on the cognitive response intensity value, the time point overall cognitive bias and the financial indicator overall cognitive bias; and performs analysis and early warning based on the bias fusion intensity coefficient.

4. The AI agent-based financial data self-adaptive optimization processing system according to claim 3, characterized in that: The data acquisition and set construction module comprises a data acquisition and set construction unit; The data acquisition and set construction unit: based on the AI agent, calls all financial indicators from the financial service platform, acquires static content data of the financial indicators and performs normalization processing, the financial indicators include market quotation data, asset price data, product data and risk indicator data, the static content data includes historical transaction data, financial indicator data and economic data of the financial indicators; aligns the static content data of the financial indicators by time point, and labels as a financial explicit data set; After user authorization, the AI agent is used to call user interaction behavior trajectory data of the financial indicators from the financial service platform backend and perform normalization processing, the interaction behavior trajectory data includes access frequency data, transaction operation data and dwell time data; The interaction behavior trajectory data is aligned by time point, and labeled as a financial implicit data set. 5.The AI agent-based financial data self-adaptive optimization processing system according to claim 4, characterized in that: The cognitive bias factor calculation module comprises a cognitive bias factor calculation unit; The cognitive bias factor calculation unit: based on the financial explicit data set and the financial implicit data set, calculates the cognitive bias factor of the user to the a-th financial indicator at the i-th time point. 6.The AI agent-based financial data self-adaptive optimization processing system according to claim 5, characterized in that: The intensity value and overall cognitive bias calculation module comprises an intensity value calculation unit and an overall cognitive bias calculation unit; The intensity value calculation unit: based on the cognitive bias factor of the user to the a-th financial indicator at the i-th time point, calculates the cognitive balance weight of the user to the a-th financial indicator at the i-th time point; Based on the cognitive balance weight of the user to the a-th financial indicator at the i-th time point and the cognitive bias factor of the user to the a-th financial indicator at the i-th time point, the cognitive response intensity value of the user to the a-th financial indicator at the i-th time point is calculated; The overall cognitive bias calculation unit: respectively acquires the cognitive response intensity value of the user to all financial indicators at the i-th time point and the cognitive response intensity value of the user to the a-th financial indicator at all time points, and calculates the time point-based cognitive response intensity mean value and the financial indicator-based cognitive response intensity mean value, respectively, which are denoted as time point overall cognitive bias and financial indicator overall cognitive bias.

7. The AI agent-based financial data self-adaptive optimization processing system according to claim 6, characterized in that: The intensity coefficient calculation and analysis warning module comprises an intensity coefficient calculation unit and an analysis warning unit; The intensity coefficient calculation unit: based on the cognitive response intensity value of the user to the a-th financial indicator at the i-th time point, the time point overall cognitive bias and the financial indicator overall cognitive bias, calculates the bias fusion intensity coefficient of the user; The analysis warning unit: if the bias fusion intensity coefficient of the user to the a-th financial indicator at the i-th time point is greater than 1 or less than 1, it is determined that the user has positive bias or negative bias to the a-th financial indicator at the i-th time point, that is, overreaction or insufficient reaction, and relevant staff are warned; all time points and all financial indicators are traversed for dynamic analysis and warning.

Images