An intelligent analysis method to improve financial decision-making efficiency

Abstract

The present invention relates to the field of data analysis technology, and specifically discloses an intelligent analysis method for improving the efficiency of financial decision-making, comprising the following steps: S1: extracting target indicators from financial information, and classifying the financial information based on the target indicators to obtain information classes; S2: preprocessing the target indicators to obtain standard indicators, generating standard points based on the standard indicators, and determining the minimum enclosing rectangle of the area formed by the standard points; S3: gridding the minimum enclosing rectangle, clustering the standard points with grid intersections and standard points as cluster centers to obtain clusters, determining the density of standard points in the clusters, screening standard clusters based on the standard point density, performing a decomposition step to update the standard clusters to obtain a target cluster; determining the center of the minimum enclosing circle of the target cluster, extracting the financial data corresponding to the center of the circle, and displaying it to the user. The present invention improves the efficiency of financial decision-making by extracting core data from massive amounts of financial data.

Description

Technical Field

[0001] The present invention relates to the field of data analysis technology, and in particular to an intelligent analysis method for improving financial decision-making efficiency. Background Art

[0002] Financial decision-making refers to the process by which decision-makers (individuals, businesses, or governments) make choices regarding capital investment, risk assumption, and expected returns, based on considerations of capital use, allocation, and management, within the context of limited resources and constraints. It is a core component of financial activity, aiming to achieve optimal allocation of funds and maximize their value.

[0003] Financial markets change rapidly, with frequently updated information. This includes not only traditional market data such as stock prices, trading volume, and volatility, but also a wide range of information, including macroeconomic indicators and industry trends. Consequently, the amount of data used for financial decision-making is enormous. Due to the sheer volume and complexity of this data, analyzing and processing this information requires significant time and resources, which can lead to delayed decisions and missed market opportunities. Furthermore, excessive information can lead to information overload, making it difficult for decision-makers to quickly and accurately extract key data, thus compromising the efficiency of financial decision-making. Summary of the Invention

[0004] The purpose of this invention is to provide an intelligent analysis method to improve the efficiency of financial decision-making and solve the following technical problems:

[0005] Due to the large amount and complexity of data, analyzing and processing this information requires a lot of time and resources, which may lead to delayed decision-making and missed market opportunities. At the same time, too much information may lead to information overload, making it difficult for decision makers to quickly and accurately extract key data, thus affecting the efficiency of financial decision-making.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] An intelligent analysis method for improving financial decision-making efficiency includes the following steps:

[0008] S1: Marking indicators that affect financial decisions as target indicators, extracting target indicators from financial information, and classifying the financial information based on the target indicators to obtain information classes, wherein the financial information of a single information class contains the same number and types of target indicators;

[0009] S2: Mark the financial information in the single information class as target information, pre-process the target indicator corresponding to the target information to obtain a standard indicator, and generate standard points (C1, C2, ..., C m ), C m represents the mth standard index, and determines the minimum circumscribed rectangle D of the area composed of the standard points;

[0010] S3: Divide the minimum circumscribed rectangle D into a grid, wherein the length and width of each grid are both preset values, set a cluster radius, cluster the standard points with the grid intersection and the standard point as the cluster center to obtain clusters, determine the density of the standard points in the clusters, select standard clusters from the clusters based on the standard point density, and execute a preset decomposition step to update the standard clusters to obtain target clusters;

[0011] The center Pyx of the minimum circumscribed circle P of the target cluster is determined, and the financial data corresponding to the center Pyx is extracted and displayed to the user to assist the user in making financial decisions.

[0012] As a further solution of the present invention: in step S3, the process of determining the standard point density in the cluster specifically includes:

[0013] Calculate the standard point density F=f / (πr 2 ), f represents the number of target points in the cluster, and r represents the radius of the minimum circumscribed circle of the area composed of the standard points.

[0014] As a further solution of the present invention: in step S3, the process of determining the target cluster specifically includes:

[0015] The standard cluster containing the same standard point is taken as the undetermined cluster, the undetermined cluster with the largest standard point density is taken as the receiving cluster, and the standard point is merged into the receiving cluster until the standard cluster does not contain the same standard point;

[0016] Determine the standard point density of the standard cluster. When the standard point density is less than a preset density threshold Fys, use the corresponding standard cluster as a decomposition cluster, and merge the standard points in the decomposition cluster into the nearest standard cluster.

[0017] When a new standard point is included in the standard cluster, the standard point density is calculated. If the standard point density is greater than or equal to the density threshold Fys, the standard point is retained. Otherwise, the standard point is removed as noise.

[0018] All standard clusters are taken as target clusters.

[0019] As a further solution of the present invention: in step S2, the process of determining the standard point composition area specifically includes:

[0020] The plane where the standard point is located is used as a reference plane, and the standard point is used as the origin to establish a plane rectangular coordinate system in the reference plane;

[0021] Determine the target quadrant according to a preset rule, determine the standard point A in the target quadrant that is closest to the origin, and connect point A and the origin with a straight line;

[0022] Repeat the above steps until all standard points are connected by straight lines to obtain a connection diagram;

[0023] The area of the minimum circumscribed rectangle of the connection graph is determined, the connection graph corresponding to the minimum area is used as the target graph, and the target graph is used as the area composed of the standard points.

[0024] As a further solution of the present invention: the process of determining the target quadrant is specifically as follows:

[0025] Let the priorities of the first, second, third, and fourth quadrants be 1, 2, 3, and 4 respectively from high to low;

[0026] According to the priority order, when there is a standard point in the first quadrant of the plane rectangular coordinate system, the first quadrant is the target quadrant; when there is no standard point in the first quadrant of the plane rectangular coordinate system, it is determined whether the second quadrant is the target quadrant;

[0027] Repeat the above steps until the target quadrant is determined.

[0028] As a further solution of the present invention: in step S3, the process of screening the standard cluster specifically includes:

[0029] The clusters are sorted in descending order according to the size of the standard point density to obtain a cluster sorting. Starting from the last cluster in the cluster sorting, the clusters are removed in sequence until the number of standard points not included in the cluster after removing a certain cluster is greater than or equal to the preset number, and the cluster at this time is used as the standard cluster.

[0030] As a further solution of the present invention: when the standard densities of two or more clusters are the same, the closer the cluster center of the cluster is to the center of the minimum circumscribed rectangle D, the higher the ranking of the corresponding cluster.

[0031] As a further solution of the present invention: in step S2, the process of determining the standard index specifically includes:

[0032] Encode non-numerical target indicators;

[0033] Standardize numerical target indicators to eliminate dimensions.

[0034] Beneficial effects of the present invention: In this solution, by extracting key indicators (target indicators) that have an important impact on financial decision-making, redundant information is preliminarily filtered, and then the target indicators are classified, and information containing the same number and type of target indicators is classified into one category, simplifying the complexity of data processing, thereby reducing the amount of data and standardizing the data structure, providing a basis for subsequent steps, avoiding processing irrelevant or redundant information, saving time and resources; thereafter, the target information is preprocessed (including encoding and standardization) to eliminate the differences between different target indicators and make the data comparable, and the preprocessed target indicators are mapped to standard points, and the distribution area of these standard points in the plane space (minimum circumscribed rectangle D) is determined, so that the complex data is visualized and structured, which is convenient for clustering and screening operations in subsequent steps, laying the foundation for the final extraction of core data; thereafter, by gridding the minimum circumscribed rectangle, the search range of the clustering operation is limited, and the efficiency is improved. Clustering efficiency is improved, and standard points are clustered based on density and distance characteristics, and standard clusters with high relevance and representativeness are screened out. Data are further layered through clustering, key information is highlighted, and information overload caused by excessive data volume is reduced. The screening and optimization of standard clusters ensure that only high-value data is retained, providing more accurate data support for subsequent decision-making. Then, the standard clusters preliminarily screened out are updated, noise data or low-density clusters are removed, and important clusters are further optimized to finally obtain the target cluster as the most valuable core data set for decision-making, further filtering noise and invalid information so that decision makers focus on high-density, high-value target clusters. Finally, the center of the minimum circumscribed circle of the target cluster is extracted as the refinement of the core data of the target cluster, and the core data is displayed to the user, intuitively providing a basis for decision-making, realizing the simplification of complex information into core points, quickly assisting users in understanding and judging, avoiding information overload, improving decision-making efficiency, and helping users grasp market opportunities. The present invention finally realizes the rapid and accurate display of key data to users through layer-by-layer filtering and extraction, thereby improving the efficiency of financial decision-making and reducing the negative impact of information overload and decision delay on efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The present invention will be further described below with reference to the accompanying drawings.

[0036] Figure 1 It is a flow chart of an intelligent analysis method for improving financial decision-making efficiency according to the present invention. DETAILED DESCRIPTION

[0037] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0038] See also Figure 1 As shown, the present invention is an intelligent analysis method for improving the efficiency of financial decision-making, comprising the following steps:

[0039] S1: Marking indicators that affect financial decisions as target indicators, extracting target indicators from financial information, and classifying the financial information based on the target indicators to obtain information classes, wherein the financial information of a single information class contains the same number and types of target indicators;

[0040] S2: Mark the financial information in the single information class as target information, pre-process the target indicator corresponding to the target information to obtain a standard indicator, and generate standard points (C1, C2, ..., C m ), C m represents the mth standard index, and determines the minimum circumscribed rectangle D of the area composed of the standard points;

[0041] S3: Divide the minimum circumscribed rectangle D into a grid, wherein the length and width of each grid are both preset values, set a cluster radius, cluster the standard points with the grid intersection and the standard point as the cluster center to obtain clusters, determine the density of the standard points in the clusters, select standard clusters from the clusters based on the standard point density, and execute a preset decomposition step to update the standard clusters to obtain target clusters;

[0042] The center Pyx of the minimum circumscribed circle P of the target cluster is determined, and the financial data corresponding to the center Pyx is extracted and displayed to the user to assist the user in making financial decisions.

[0043] It should be noted that by extracting key indicators (target indicators) that have a significant impact on financial decision-making, redundant information is preliminarily filtered, and then the target indicators are classified, and information containing the same number and type of target indicators is grouped into one category, the complexity of data processing is simplified, thereby reducing the amount of data and standardizing the data structure, providing a basis for subsequent steps, avoiding processing irrelevant or redundant information, saving time and resources; then, the target information is preprocessed (including encoding and standardization) to eliminate the differences between different target indicators and make the data comparable, and the preprocessed target indicators are mapped to standard points, and the distribution area of these standard points in the plane space is determined (minimum enclosing rectangle D), so that the complex data is visualized and structured, which is convenient for clustering and screening operations in subsequent steps, laying the foundation for the final extraction of core data; then, by gridding the minimum enclosing rectangle, the search range of the clustering operation is limited, the clustering efficiency is improved, and the standard points are clustered based on density and distance characteristics to screen out those with high The standard clusters of relevance and representativeness are clustered to further stratify the data, highlight key information, and reduce information overload caused by excessive data volume. The screening and optimization of standard clusters ensure that only high-value data are retained, providing more accurate data support for subsequent decision-making; then, the initially screened standard clusters are updated, noise data or low-density clusters are removed, and important clusters are further optimized to finally obtain the target cluster as the most valuable core data set for decision-making, further filtering out noise and invalid information so that decision makers can focus on high-density, high-value target clusters; finally, the center of the minimum circumscribed circle of the target cluster is extracted as the refinement of the core data of the target cluster, and the core data is displayed to the user, providing an intuitive basis for decision-making, simplifying complex information into core points, quickly assisting users in understanding and judgment, avoiding information overload, improving decision-making efficiency, and helping users seize market opportunities; among them, when the coordinates of the center of the circle contain a value that is not an integer, the standard point G closest to the circle is determined, and the financial data on the standard point G is extracted and displayed to the user.

[0044] In another preferred embodiment of the present invention, in step S3, the process of determining the standard point density in the cluster specifically includes: calculating the standard point density F=f / (πr 2 ), f represents the number of target points in the cluster, and r represents the radius of the minimum circumscribed circle of the area composed of the standard points.

[0045] In another preferred embodiment of the present invention, in step S3, the process of determining the target cluster specifically includes:

[0046] The standard cluster containing the same standard point is taken as the undetermined cluster, the undetermined cluster with the largest standard point density is taken as the receiving cluster, and the standard point is merged into the receiving cluster until the standard cluster does not contain the same standard point;

[0047] Determine the standard point density of the standard cluster. When the standard point density is less than a preset density threshold Fys, use the corresponding standard cluster as a decomposition cluster, and merge the standard points in the decomposition cluster into the nearest standard cluster.

[0048] When a new standard point is included in the standard cluster, the standard point density is calculated. If the standard point density is greater than or equal to the density threshold Fys, the standard point is retained. Otherwise, the standard point is removed as noise.

[0049] All standard clusters are taken as target clusters.

[0050] It can be understood that the standard clusters containing the same standard point are classified as pending clusters to avoid repeated processing of the same data point. By selecting the pending cluster with the largest density as the receiving cluster, the most representative cluster structure can be retained first, the data integration efficiency can be improved, and the final target cluster can be ensured to have a high degree of information concentration and availability; when the density of the standard cluster is less than the preset density threshold, it is decomposed into decomposition clusters, and its points are merged into the nearest standard cluster, which can avoid the low-density cluster from containing noise points or invalid data, thereby reducing the quality of the target cluster. Dynamic adjustment of the cluster structure can effectively eliminate the interference of low-density clusters, assign important points to high-correlation clusters, and further enhance the target cluster's ability to describe key information; all standard clusters that have undergone the above processing are used as target clusters to avoid the characteristic differences between different information being masked and the loss of the description of data diversity. Among them, the standard point density of the standard cluster can refer to the cluster cluster, which will not be repeated here.

[0051] In another preferred embodiment of the present invention, in step S2, the process of determining the standard point composition area specifically includes:

[0052] The plane where the standard point is located is used as a reference plane, and the standard point is used as the origin to establish a plane rectangular coordinate system in the reference plane;

[0053] Determine the target quadrant according to a preset rule, determine the standard point A in the target quadrant that is closest to the origin, and connect point A and the origin with a straight line;

[0054] Repeat the above steps until all standard points are connected by straight lines to obtain a connection diagram;

[0055] The area of the minimum circumscribed rectangle of the connection graph is determined, the connection graph corresponding to the minimum area is used as the target graph, and the target graph is used as the area composed of the standard points.

[0056] In another preferred embodiment of the present invention, the process of determining the target quadrant is specifically as follows:

[0057] Let the priorities of the first, second, third, and fourth quadrants be 1, 2, 3, and 4 respectively from high to low;

[0058] According to the priority order, when there is a standard point in the first quadrant of the plane rectangular coordinate system, the first quadrant is the target quadrant; when there is no standard point in the first quadrant of the plane rectangular coordinate system, it is determined whether the second quadrant is the target quadrant;

[0059] Repeat the above steps until the target quadrant is determined.

[0060] In another preferred embodiment of the present invention, in step S3, the process of screening the standard cluster specifically includes:

[0061] The clusters are sorted in descending order according to the size of the standard point density to obtain a cluster sorting. Starting from the last cluster in the cluster sorting, the clusters are removed in sequence until the number of standard points not included in the cluster after removing a certain cluster is greater than or equal to the preset number, and the cluster at this time is used as the standard cluster.

[0062] It is understandable that a large number of clusters may be generated during the clustering process, some of which have low density or contain low information value and do not have sufficient analytical significance. Retaining all clusters will lead to redundancy or increased noise in the target cluster, affecting subsequent data processing and analysis. Density is an important indicator for measuring the quality of clusters. The higher the density, the more concentrated the distribution of points in the cluster, which means that the correlation between these points is stronger. By sorting in descending order, we can quickly focus on high-density clusters and gradually eliminate them starting from low-density clusters to avoid interference with high-value clusters from the beginning. The preset number can be set according to actual conditions or by professionals, such as 0, so that all standard points are retained.

[0063] In another preferred embodiment of the present invention, when two or more clusters have the same standard density, the closer the cluster center is to the center of the minimum circumscribed rectangle D, the higher the ranking of the corresponding cluster.

[0064] In another preferred embodiment of the present invention, in step S2, the process of determining the standard index specifically includes:

[0065] Encode non-numerical target indicators;

[0066] Standardize numerical target indicators to eliminate dimensions.

[0067] The above is a detailed description of an embodiment of the present invention. However, the content described is only a preferred embodiment of the present invention and should not be considered to limit the scope of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. An intelligent analysis method for improving financial decision-making efficiency, characterized by: The following steps are involved: S1: Marking indicators that affect financial decisions as target indicators, extracting target indicators from financial information, and classifying the financial information based on the target indicators to obtain information classes, wherein the financial information of a single information class contains the same number and types of target indicators; S2: Mark the financial information in the single information class as target information, pre-process the target indicator corresponding to the target information to obtain a standard indicator, and generate standard points (C1, C2, ..., C m ), C m represents the mth standard index, and determines the minimum circumscribed rectangle D of the area composed of the standard points; S3: Divide the minimum circumscribed rectangle D into a grid, wherein the length and width of each grid are both preset values, set a cluster radius, cluster the standard points with the grid intersection and the standard point as the cluster center to obtain clusters, determine the density of the standard points in the clusters, select standard clusters from the clusters based on the standard point density, and execute a preset decomposition step to update the standard clusters to obtain target clusters; Determine the center Pyx of the minimum circumscribed circle P of the target cluster, extract the financial data corresponding to the center Pyx and display it to the user to assist the user in making financial decisions; The process of determining the target cluster specifically includes: The standard cluster containing the same standard point is taken as the undetermined cluster, the undetermined cluster with the largest standard point density is taken as the receiving cluster, and the standard point is merged into the receiving cluster until the standard cluster does not contain the same standard point; Determine the standard point density of the standard cluster. When the standard point density is less than a preset density threshold Fys, use the corresponding standard cluster as a decomposition cluster, and merge the standard points in the decomposition cluster into the nearest standard cluster. When a new standard point is included in the standard cluster, the standard point density is calculated. If the standard point density is greater than or equal to the density threshold Fys, the standard point is retained. Otherwise, the standard point is removed as noise. All standard clusters are used as target clusters; The process of determining the standard point composition area specifically includes: The plane where the standard point is located is used as a reference plane, and the standard point is used as the origin to establish a plane rectangular coordinate system in the reference plane; Determine the target quadrant according to a preset rule, determine the standard point A in the target quadrant that is closest to the origin, and connect point A and the origin with a straight line; Repeat the above steps until all standard points are connected by straight lines to obtain a connection diagram; The area of the minimum circumscribed rectangle of the connection graph is determined, the connection graph corresponding to the minimum area is used as the target graph, and the target graph is used as the area composed of the standard points.

2. The intelligent analysis method for improving financial decision-making efficiency according to claim 1, characterized in that: In step S3, the process of determining the standard point density in the cluster specifically includes: Calculate the standard point density F=f / (πr 2 ), f represents the number of target points in the cluster, and r represents the radius of the minimum circumscribed circle of the area composed of the standard points.

3. The intelligent analysis method for improving financial decision-making efficiency according to claim 1, characterized in that: The process of determining the target quadrant is as follows: Let the priorities of the first, second, third, and fourth quadrants be 1, 2, 3, and 4 respectively from high to low; According to the priority order, when there is a standard point in the first quadrant of the plane rectangular coordinate system, the first quadrant is the target quadrant; when there is no standard point in the first quadrant of the plane rectangular coordinate system, it is determined whether the second quadrant is the target quadrant; Repeat the above steps until the target quadrant is determined.

4. The intelligent analysis method for improving financial decision-making efficiency according to claim 1, characterized in that: In step S3, the process of screening the standard cluster specifically includes: The clusters are sorted in descending order according to the size of the standard point density to obtain a cluster sorting. Starting from the last cluster in the cluster sorting, the clusters are removed in sequence until the number of standard points not included in the cluster after removing a certain cluster is greater than or equal to the preset number, and the cluster at this time is used as the standard cluster.

5. The intelligent analysis method for improving financial decision-making efficiency according to claim 1, characterized in that: When the standard densities of two or more clusters are the same, the closer the cluster center is to the center of the minimum circumscribed rectangle D, the higher the corresponding cluster ranking.

6. The intelligent analysis method for improving financial decision-making efficiency according to claim 1, characterized in that: In step S2, the process of determining the standard index specifically includes: Encode non-numerical target indicators; Standardize numerical target indicators to eliminate dimensions.

Images