Financial data analysis method and system based on artificial intelligence

Abstract

The invention belongs to the technical field of financial analysis, and discloses a financial data analysis method and system based on artificial intelligence, and the system comprises a data receiving module, a feature extraction stage, a training model stage, a detection module, a prediction module, and a display module. The financial data of stock exchange, financial institutions, social media and macroeconomic data are collected through the data receiving module, the data useful for prediction and classification tasks are substituted into the learning model, the model learns rules and modes in the data, and then the model is monitored through the detection module, so that the prediction and classification tasks can be predicted and classified. And when the model detection does not pass, the model is traced back into the data collection module for further learning so as to ensure that an operation model does not have an over-fitting or under-fitting condition, so that future financial data can be predicted through the model, and an analysis result is visually displayed in forms of charts, reports and the like, so that a user can conveniently understand and use the analysis result.

Description

Technical Field

[0001] The present invention relates to the technical field of financial analysis, and specifically to a financial data analysis method and system based on artificial intelligence. Background Art

[0002] In addition to the general characteristics of data, financial data also has some of its own characteristics: extensiveness, comprehensiveness, reliability, and continuity; the particularity of financial data makes the processing of financial data also have its special points and special requirements. Its input review is more stringent, the storage capacity is larger, the network transmission is more extensive, and the data maintenance is more frequent;

[0003] Existing financial data analysis systems usually use built-in operation models to predict financial data, but overfitting or underfitting may occur when using the operation models, resulting in errors in the prediction results obtained by the financial data system. Summary of the Invention

[0004] In view of the deficiencies of the prior art, the present invention provides a financial data analysis method and system based on artificial intelligence, which solves the problems raised in the above background art.

[0005] To solve the above technical problems, according to one aspect of the present invention, more specifically, a financial data analysis method and system based on artificial intelligence includes a data receiving module, a detection module, a prediction module, and a display module, and is characterized in that: the data receiving module is used to collect financial data. Among them, the artificial intelligence system collects financial data from stock exchanges, financial institutions, social media, and macroeconomic data through API interfaces, data crawlers, and data warehouses, trains a model based on the collected financial data, and applies it to the artificial intelligence system;

[0006] The detection module: is used to detect the calculation results of the model;

[0007] The prediction module: predicts future financial data through the model and converts the prediction results into investment recommendations, risk management strategies, or investment strategies;

[0008] The display module: is used to intuitively display the analysis results in the form of charts, reports, etc., for easy understanding and use by users.

[0009] Furthermore, it includes the following steps:

[0010] S1. Collect financial data from stock exchanges, financial institutions, social media, and macroeconomic data through API interfaces, data crawlers, and data warehouses, and clean, denoise, and format the collected data to ensure data quality;

[0011] S2. Extract features meaningful for financial analysis from the collected data, select the features most useful for prediction and classification tasks, and substitute them into the learning model to enable the model to learn the patterns and regularities in the data;

[0012] S3. Monitor the model through the detection module, and when the model fails the detection, trace the model back to the data collection module for further learning;

[0013] S4. Use the model to predict future financial data and convert the prediction results into investment recommendations, risk management strategies, or investment strategies;

[0014] S5. Intuitively display the analysis results in the form of charts, reports, etc., for easy understanding and use by users.

[0015] Furthermore, the calculation formula for the model result detection by the detection module is:

[0016]

[0017] In the formula, R is the error index of the model structure; a is the number of positive class samples correctly predicted as positive by the model; b is the number of positive class samples incorrectly predicted as negative by the model; c is the number of negative class samples incorrectly predicted as positive by the model; d is the number of samples correctly predicted as negative by the model.

[0018] Furthermore, a data cleaning module is provided inside the data receiving module to perform preprocessing operations such as data cleaning, denoising, and standardization on the data to improve data quality and analysis effect.

[0019] Furthermore, the data cleaning module has functions of data deduplication, missing value processing, and outlier detection, where:

[0020] Data deduplication: used to delete duplicate data to ensure data uniqueness;

[0021] Missing value processing: fill in missing values by interpolation and regression methods, and at the same time, select to delete records with more missing values according to the data characteristics;

[0022] Outlier detection: identify and process outliers caused by data entry errors and equipment failures.

[0023] Furthermore, the financial data analysis system model uses historical data to train the model, enables the model to learn the patterns and regularities in the data, and optimizes the model performance and reduces the risk of overfitting or underfitting by adjusting model parameters, using regularization techniques, and cross-validation methods.

[0024] Further, when the data receiving module transmits data, it encrypts sensitive data and at the same time allows calculations and analyses to be performed on the encrypted data, achieving data analysis while protecting data privacy.

[0025] The beneficial effects of a financial data analysis method and system based on artificial intelligence according to the present invention are as follows:

[0026] The present invention collects financial data from stock exchanges, financial institutions, social media, and macroeconomic data through a data receiving module, and substitutes the data useful for prediction and classification tasks into a learning model to allow the model to learn the rules and patterns in the data. Then, the model is monitored by a detection module, and when the model fails the detection, the model is traced back to the data collection module for further learning to ensure that the operation model does not overfit or underfit, so as to predict future financial data through the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The present invention will be further described in detail below with reference to the drawings and specific implementation methods.

[0028] Figure 1 It is a schematic structural diagram of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0029] The present invention will be described in detail below with reference to the drawings and embodiments. It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments may be combined with each other.

[0030] As Figure 1 shown, according to one aspect of the present invention, there is provided a financial data analysis method and system based on artificial intelligence, including a data receiving module, a detection module, a prediction module, and a display module, characterized in that: the data receiving module is used to collect financial data, wherein the artificial intelligence system collects financial data from stock exchanges, financial institutions, social media, and macroeconomic data through API interfaces, data crawlers, and data warehouses, trains a model according to the collected financial data, and applies it to the artificial intelligence system;

[0031] Detection module: used to detect the model calculation results;

[0032] Prediction module: predicts future financial data through the model and converts the prediction results into investment recommendations, risk management strategies, or investment strategies;

[0033] Display module: used to intuitively display the analysis results in the form of charts, reports, etc., for easy understanding and use by users.

[0034] In this embodiment, the following steps are included:

[0035] S1. Collect financial data from stock exchanges, financial institutions, social media, and macroeconomic data through API interfaces, data crawlers, and data warehouses, and clean, denoise, and format the collected data to ensure data quality;

[0036] S2. Extract features meaningful for financial analysis from the collected data, select the most useful features for prediction and classification tasks, and substitute them into the learning model to let the model learn the patterns and regularities in the data;

[0037] S3. Monitor the model through the detection module, and when the model fails the detection, trace the model back to the data collection module for further learning;

[0038] S4. Predict future financial data through the model and convert the prediction results into investment recommendations, risk management strategies, or investment strategies;

[0039] S5. Intuitively display the analysis results in the form of charts, reports, etc., for easy understanding and use by users.

[0040] In this embodiment, the calculation formula for the detection module to detect the model results is:

[0041]

[0042] In the formula, R is the error index of the model structure; a is the number of positive class samples correctly predicted as positive by the model; b is the number of positive class samples incorrectly predicted as negative by the model; c is the number of negative class samples incorrectly predicted as positive by the model; d is the number of samples correctly predicted as negative by the model.

[0043] In this embodiment, a data cleaning module is provided inside the data receiving module to perform preprocessing operations such as data cleaning, denoising, and standardization on the data to improve data quality and analysis effect.

[0044] In this embodiment, the data cleaning module has functions of data deduplication, missing value processing, and outlier detection, where:

[0045] Data deduplication: used to delete duplicate data to ensure data uniqueness;

[0046] Missing value processing: fill in missing values through interpolation and regression methods. At the same time, select records with more missing values for deletion according to the data characteristics;

[0047] Outlier detection: identify and process outliers caused by data entry errors and equipment failures.

[0048] In this embodiment, the financial data analysis system model is trained using historical data to enable the model to learn the patterns and rules in the data. By adjusting the model parameters, using regularization techniques, and cross-validation methods, the model performance is optimized, and the risks of overfitting or underfitting are reduced.

[0049] In this embodiment, when the data receiving module transmits data, it encrypts sensitive data and at the same time allows calculations and analyses to be performed on the encrypted data, achieving data analysis while protecting data privacy.

[0050] Embodiment

[0051] When the detection module detects the operation model:

[0052] 100 groups of financial data are sampled, where:

[0053] The number (a) of positive class samples correctly predicted as positive by the model is: 95 groups;

[0054] The number (b) of positive class samples incorrectly predicted as negative by the model is: 3 groups;

[0055] The number (c) of negative class samples incorrectly predicted as positive by the model is: 1 group;

[0056] The number (d) of negative class samples correctly predicted as negative by the model is: 1 group;

[0057] Substituting the above data into the detection calculation formula gives:

[0058]

[0059] It is obtained that 0.97 < 1, and this group of financial data analysis models is normal and can be used for subsequent financial predictions.

[0060] Certainly, the above description is not a limitation of the present invention, and the present invention is not limited to the above examples either. Changes, modifications, additions, or substitutions made by those of ordinary skill in the art within the scope of the essence of the present invention also belong to the protection scope of the present invention.

Claims

1. A financial data analysis system based on artificial intelligence, comprising a data receiving module, a detection module, a prediction module and a display module, characterized in that: The data receiving module is used to collect financial data, wherein the artificial intelligence system collects financial data from stock exchanges, financial institutions, social media, and macroeconomic data through API interfaces, data crawlers, and data warehouses, and trains a model based on the collected financial data and applies it to the artificial intelligence system; Detection module: used to detect model calculation results; Prediction module: predicts future financial data through models and converts the prediction results into investment advice, risk management strategies or investment strategies; Display module: used to intuitively display the analysis results in the form of charts, reports, etc., to facilitate user understanding and use.

2. The financial data analysis method based on artificial intelligence according to claim 1, characterized in that: The following steps are involved: S1. Collect financial data from stock exchanges, financial institutions, social media, and macroeconomic data through API interfaces, data crawlers, and data warehouses, and clean, denoise, and format the collected data to ensure data quality; S2. Extract features that are meaningful for financial analysis from the collected data, select the most useful features for prediction and classification tasks, substitute them into the learning model, and let the model learn the rules and patterns in the data; S3, monitoring the model through the detection module, and when the model fails the detection, tracing the model back to the data collection module for further learning; S4. Use models to predict future financial data and convert the prediction results into investment advice, risk management strategies or investment strategies; S5. Display the analysis results intuitively in the form of charts, reports, etc. to facilitate user understanding and use.

3. The financial data analysis system based on artificial intelligence according to claim 1, characterized in that: The calculation formula for the detection module to detect the model results is: Where R is the error index of the model structure; a is the number of positive samples correctly predicted as positive by the model; b is the number of positive samples incorrectly predicted as negative by the model; c is the number of negative samples incorrectly predicted as positive by the model; and d is the number of samples correctly predicted as negative by the model.

4. The financial data analysis system based on artificial intelligence according to claim 1, characterized in that: The data receiving module is internally provided with a data cleaning module to perform pre-processing operations such as cleaning, denoising, and standardization on the data to improve the data quality and analysis effect.

5. The financial data analysis method and system based on artificial intelligence according to claim 4, characterized in that: The data cleaning module has the functions of data deduplication, missing value processing and outlier detection, among which: Data deduplication: used to delete duplicate data to ensure data uniqueness; Missing value processing: fill in missing values ​​through interpolation and regression methods, and delete records with more missing values ​​according to data characteristics; Outlier detection: Identify and handle outliers caused by data entry errors and equipment failures.

6. The financial data analysis system based on artificial intelligence according to claim 1, characterized in that: The financial data analysis system model uses historical data to train the model, allowing the model to learn the laws and patterns in the data, and optimizes model performance and reduces the risk of overfitting or underfitting by adjusting model parameters, using regularization techniques, and cross-validation methods.

7. The financial data analysis system based on artificial intelligence according to claim 1, characterized in that: When the data receiving module transmits data, it encrypts the sensitive data and allows calculation and analysis on the encrypted data, thereby protecting data privacy and realizing data analysis.

Images