Financing data analysis method and system

Abstract

The invention provides a financing data analysis method and system, and the method comprises the steps: calling original financing data from a database, the original financing data comprising numeric data, classification data and time data; performing quantitative feature extraction on the numeric data, wherein the numeric data comprises financing amount, historical loan times, repayment amount and trade exchange data; performing qualitative feature extraction on the classified data, wherein the classified data comprises the enterprise industry, the enterprise scale and the financing purpose; and performing time feature extraction on the time data, and decomposing the time data in the time feature extraction process. According to the scheme, the original financing data can be collected from the original database, the analyzed data corresponding to the multiple items can be obtained by conducting data analysis on the multiple items of original financing data, the analyzed data can be displayed, and the efficient tactics of the financing data is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and in particular to a financing data analysis method and system. Background Art

[0002] As an important indicator of the company's financial health and market competitiveness, corporate financing data is of great importance. These data not only reflect the company's current and past financing activities, but also indicate the company's future development direction and potential.

[0003] First, financing data is an important basis for enterprises to make strategic decisions. Through in-depth analysis of financing data, enterprises can clearly understand key information such as their capital structure, financing costs and financing efficiency. This information helps enterprise decision makers judge whether the current financing strategy is effective and whether it is necessary to adjust the financing method or channel to better meet the capital needs of the enterprise. At the same time, financing data can also help enterprises predict future capital needs, so as to plan financing plans in advance and ensure the stable operation of the enterprise's capital chain. Secondly, financing data is also crucial to enhance the market reputation and image of enterprises. A company with a good financing record can often gain the trust and favor of investors more easily. This not only helps enterprises reduce financing costs, but also brings more investment opportunities and partners to enterprises.

[0004] Therefore, by showing good financing data, companies can attract more investors and partners, thereby expanding their business areas and increasing their market share. At the same time, financing data can also bring opportunities for financing innovation to companies, such as issuing bonds, equity financing and other new financing methods, further broadening the financing channels of companies and reducing financing costs. However, the existing technology lacks effective analysis methods for financing data, making it difficult to achieve efficient display of financing data. Summary of the invention

[0005] In view of this, an embodiment of the present invention provides a financing data analysis method to eliminate or improve one or more defects existing in the prior art.

[0006] One aspect of the present invention provides a financing data analysis method, the method comprising the steps of:

[0007] Retrieving original financing data from a database, wherein the original financing data includes numerical data, classification data, and time data;

[0008] Performing quantitative feature extraction on the numerical data, wherein the numerical data includes financing amount, historical borrowing times, repayment amount and trade data;

[0009] Extracting qualitative features from the classified data, wherein the classified data includes enterprise industry, enterprise size, and financing purpose;

[0010] Time feature extraction is performed on the time data, and the time data is decomposed during the time feature extraction process.

[0011] By adopting the above scheme, this scheme can collect original financing data from the original database, and obtain analyzed data corresponding to multiple projects by performing data analysis on multiple original financing data respectively. This scheme can display multiple analyzed data to achieve efficient tactics for financing data.

[0012] In some embodiments of the present invention, the method further includes: integrating the data obtained through quantitative feature extraction, qualitative feature extraction, and qualitative feature extraction, inputting the data obtained through the data integration into a pre-trained machine learning model, and outputting the financing risk value through the pre-trained machine learning model.

[0013] In some embodiments of the present invention, in the step of integrating the data obtained by quantitative feature extraction, qualitative feature extraction and qualitative feature extraction:

[0014] In the process of extracting the quantitative features, the mean and standard deviation of the financing amounts in multiple years are calculated.

[0015] In some embodiments of the present invention, in the step of integrating the data obtained by quantitative feature extraction, qualitative feature extraction and qualitative feature extraction:

[0016] In the process of extracting qualitative features from the categorical data, label encoding or one-hot encoding is used to extract features from the categorical data.

[0017] In some embodiments of the present invention, in the step of integrating the data obtained by quantitative feature extraction, qualitative feature extraction and qualitative feature extraction:

[0018] In the process of extracting time features from the classified data, the financing time data is decomposed into the year, quarter and month.

[0019] In some embodiments of the present invention, in the step of integrating the data obtained by quantitative feature extraction, qualitative feature extraction and qualitative feature extraction:

[0020] In the process of extracting time features from the classified data, the financing frequency is calculated based on the time points of multiple financing time data.

[0021] In some embodiments of the present invention, the machine learning model includes a first-stage processing model and a second-stage processing model, the first-stage processing model includes sub-models corresponding to quantitative features, qualitative feature extraction and time features, the data obtained by data integration is input into the pre-trained machine learning model, and in the step of outputting the financing risk value through the pre-trained machine learning model, the extracted quantitative features, qualitative feature extraction and time features are respectively input into the pre-trained sub-models, the outputs of multiple sub-models are combined and input into the second-stage processing model, and the financing risk value is output through the second-stage processing model.

[0022] In some embodiments of the present invention, the method further comprises:

[0023] Extracting financial indicator data from the enterprise's financial statements, wherein the financial indicator data includes current assets, current liabilities, total liabilities and total assets;

[0024] Calculate the current ratio based on the current assets and current liabilities;

[0025] The debt-to-asset ratio is calculated based on the total liabilities and total assets.

[0026] In some embodiments of the present invention, in the step of calculating the current ratio based on the current assets and current liabilities, current assets / current liabilities are calculated as the current ratio; total liabilities / total assets are calculated as the debt-to-asset ratio.

[0027] The second aspect of the present invention also provides a financing data analysis system, which includes a computer device, the computer device includes a processor and a memory, the memory stores computer instructions, the processor is used to execute the computer instructions stored in the memory, and when the computer instructions are executed by the processor, the system implements the steps implemented by the method described above.

[0028] The third aspect of the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps implemented by the aforementioned financing data analysis method.

[0029] Additional advantages, purposes, and features of the present invention will be described in part in the following description, and will become apparent to those skilled in the art after studying the following, or may be learned from the practice of the present invention. The purposes and other advantages of the present invention can be specifically pointed out and obtained in the specification and the accompanying drawings.

[0030] Those skilled in the art will appreciate that the objectives and advantages that can be achieved with the present invention are not limited to the above specific description, and the above and other objectives that can be achieved by the present invention will be more clearly understood from the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The drawings described herein are used to provide a further understanding of the present invention, constitute a part of this application, and do not constitute a limitation of the present invention.

[0032] Figure 1 A schematic diagram of an implementation of the financing data analysis method of the present invention;

[0033] Figure 2 A schematic diagram of the processing architecture of the financing data analysis method of the present invention;

[0034] Figure 3 This is a schematic diagram of the data processing flow of the financing data analysis method of the present invention. DETAILED DESCRIPTION

[0035] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments and the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0036] It should also be noted that, in order to avoid obscuring the present invention due to unnecessary details, only structures and / or processing steps closely related to the solutions according to the present invention are shown in the accompanying drawings, while other details that are not closely related to the present invention are omitted.

[0037] Financing data analysis and post-loan early warning systems are two important links in the financial field. Through the collection and analysis of financing data, the financing status, operational efficiency and future development potential of enterprises can be evaluated. At the same time, potential risks and problems can be discovered in time and early warning can be issued. It provides strong data support for corporate financing decisions and loan risk management.

[0038] The existing financing data analysis and post-loan early warning system is mainly based on big data and machine learning technology. It collects and analyzes multi-dimensional information such as the company's financial data, operating conditions, market trends, etc., evaluates the company's repayment ability and financing risks, and provides decision support for financial institutions. The big data model is used to monitor the operating conditions of the loan company in real time. Once an abnormality or potential risk is found, the early warning mechanism is immediately triggered to notify the financial institution to take corresponding measures in a timely manner.

[0039] Financing data analysis and post-loan early warning systems have significant advantages in improving risk management efficiency, enhancing risk identification capabilities, and optimizing credit resource allocation. However, they also have obvious problems in terms of data acquisition difficulty, model updating and maintenance, technical barriers and costs, as well as dependence and limitations.

[0040] like Figure 1-3As shown, the present invention proposes a financing data analysis method, the steps of which include:

[0041] Step S100, calling original financing data from a database, wherein the original financing data includes numerical data, classification data and time data;

[0042] In the specific implementation process, the database may be a relational database, a non-relational database, a data warehouse, and a time series database;

[0043] Relational database management systems (RDBMS) include:

[0044] Oracle, Oracle has high data integrity and security, and supports complex transaction processing and queries; Application scenarios: commonly used in the core business systems of large enterprises, such as database management in the fields of finance, telecommunications, etc.

[0045] SQL Server, SQL Server is seamlessly integrated with the Windows system and is highly easy to use; Application scenarios: widely used in database management of small and medium-sized enterprises, as well as business systems related to the Microsoft technology stack.

[0046] MySQL, MySQL is an open source relational database with high performance and scalability; Application scenarios: suitable for Web applications, database management of small and medium-sized enterprises, and occasions where open source solutions are needed.

[0047] DB2, DB2 is IBM's commercial database product, supports multiple operating systems, and has powerful data management and analysis capabilities; Application scenarios: commonly used in database management of large enterprises, government agencies and scientific research institutions.

[0048] Non-relational databases (NoSQL) include:

[0049] MongoDB, MongoDB is a document-oriented database that supports flexible data models and is suitable for large-scale data storage and distributed computing; Application scenarios: commonly used for database management in content management, social networks, the Internet of Things and other fields.

[0050] Cassandra, Cassandra is a distributed database with the characteristics of high availability and no single point of failure; Application scenarios: Suitable for occasions that require high availability and distributed storage, such as cloud computing, big data processing and other fields.

[0051] The data warehouse includes:

[0052] Teradata is a data warehouse system specially designed for storing and analyzing large amounts of data, supporting enterprise decision-making and business intelligence analysis; Application scenarios: commonly used in the construction of data warehouses in large enterprises, financial institutions and scientific research institutions to support data analysis and decision support.

[0053] IBM DB2 Warehouse, IBM DB2 Warehouse is a data warehouse solution provided by IBM, built on DB2 database technology, supporting high-performance data storage and analysis; Application scenarios: Suitable for occasions that require high-performance data analysis and decision support, such as retail, manufacturing and other fields.

[0054] Time series databases include:

[0055] InfluxDB, InfluxDB is a database system specifically used to store and process time series data, with high performance and scalability; Application scenarios: commonly used in database management in the fields of financial transactions, Internet of Things, sensor data, etc. to support the analysis and processing of time series data.

[0056] OpenTSDB, OpenTSDB is an open source time series database system built on Hadoop and HBase technology, suitable for the storage and analysis of large-scale time series data; Application scenarios: Suitable for occasions where large-scale time series data needs to be processed, such as Internet monitoring, smart grid and other fields.

[0057] With the above solution, the data warehouse first collects financing-related data from the financing business data source, including financing enterprise data, transaction data, trade background, credit data, etc. Then, the data is cleaned and integrated. Cleaning and integration mainly removes duplications of various business data, unifies the format, and fills in necessary missing values ​​to improve the quality and reliability of the data and provide accurate basic data for subsequent analysis.

[0058] In the specific implementation process, the process of preprocessing the data includes:

[0059] The raw data is cleaned, sorted, normalized, and other operations are performed to improve the quality and consistency of the data;

[0060] Select the most relevant and representative features from the original features to reduce redundancy and improve model performance. Feature selection methods include filtering, packaging, and embedding;

[0061] Depending on the feature extraction method chosen, useful features are extracted from the raw data. This process may involve mathematical transformation, signal processing, or deep learning techniques.

[0062] The extracted features are evaluated to determine whether they have a positive impact on model performance. Evaluation methods include cross-validation and model comparison.

[0063] Optimize the features based on the evaluation results, including adjusting the feature extraction method and adding new features.

[0064] Step S200, performing quantitative feature extraction on the numerical data, the numerical data including financing amount, historical borrowing times, repayment amount and trade data;

[0065] In addition to using the original values ​​directly, this solution also extracts quantitative features through basic statistics and ratio calculations. For example, for the amount of financing of an enterprise, the mean and standard deviation of the amount of financing in recent years can be calculated to reflect the stability of its financing.

[0066] Step S300, extracting qualitative features from the classified data, wherein the classified data includes enterprise industry, enterprise scale and financing purpose;

[0067] Step S400, extracting time features from the time data, and decomposing the time data during the time feature extraction process.

[0068] By adopting the above scheme, this scheme can collect original financing data from the original database, and obtain analyzed data corresponding to multiple projects by performing data analysis on multiple original financing data respectively. This scheme can display multiple analyzed data to achieve efficient tactics for financing data.

[0069] In some embodiments of the present invention, the method further includes: integrating the data obtained through quantitative feature extraction, qualitative feature extraction, and qualitative feature extraction, inputting the data obtained through the data integration into a pre-trained machine learning model, and outputting the financing risk value through the pre-trained machine learning model.

[0070] In some embodiments of the present invention, in the step of integrating the data obtained by quantitative feature extraction, qualitative feature extraction and qualitative feature extraction:

[0071] In the process of extracting the quantitative features, the mean and standard deviation of the financing amounts in multiple years are calculated.

[0072] Using the above scheme, the stability of its financing is reflected by the mean and standard deviation of the financing amount.

[0073] In some embodiments of the present invention, in the step of integrating the data obtained by quantitative feature extraction, qualitative feature extraction and qualitative feature extraction:

[0074] In the process of extracting qualitative features from the categorical data, label encoding or one-hot encoding is used to extract features from the categorical data.

[0075] In some embodiments of the present invention, in the step of integrating the data obtained by quantitative feature extraction, qualitative feature extraction and qualitative feature extraction:

[0076] In the process of extracting time features from the classified data, the financing time data is decomposed into the year, quarter and month.

[0077] In some embodiments of the present invention, in the step of integrating the data obtained by quantitative feature extraction, qualitative feature extraction and qualitative feature extraction:

[0078] In the process of extracting time features from the classified data, the financing frequency is calculated based on the time points of multiple financing time data.

[0079] During the specific implementation process, the financing frequency is calculated through the time difference to reflect the time trend and financing needs of customers.

[0080] During the specific implementation process, some important financial indicators are extracted based on the company's financial statements. The company's financial leverage and risk are measured by calculating the debt-to-asset ratio (total liabilities / total assets); and the repayment capacity is reflected by calculating the current ratio (current assets / current liabilities).

[0081] Using the above solution, feature extraction and modeling are important links in building a financing data analysis and post-loan early warning system. The purpose of feature extraction is to extract financing-related features from the original data.

[0082] In some embodiments of the present invention, the machine learning model includes a first-stage processing model and a second-stage processing model, the first-stage processing model includes sub-models corresponding to quantitative features, qualitative feature extraction and time features, the data obtained by data integration is input into the pre-trained machine learning model, and in the step of outputting the financing risk value through the pre-trained machine learning model, the extracted quantitative features, qualitative feature extraction and time features are respectively input into the pre-trained sub-models, the outputs of multiple sub-models are combined and input into the second-stage processing model, and the financing risk value is output through the second-stage processing model.

[0083] The model used in the first-stage processing model or the second-stage processing model of this solution can be a decision tree, random forest or support vector machine, etc., or a deep learning model such as a neural network can be used, which has advantages in processing large-scale data and complex pattern recognition to build a high-precision prediction model.

[0084] In the specific implementation process, model training, verification and testing are carried out according to the selected model type. Parameters are estimated through training data, model performance is evaluated through verification data, and the generalization ability of the model is tested through test data. In this process, we need to continuously adjust the parameters and structure of the model to optimize the prediction accuracy of the model. Finally, the intelligent decision engine integrates the features, models and strategies to form a user scorecard, and predicts risk data for early warning.

[0085] Using the above solution, model design and training optimization

[0086] 1. Clarify the objectives of the model, such as predicting customer delinquency probability, identifying fraudulent behavior, etc.

[0087] 2. Select the appropriate model type based on the model objectives and data characteristics, mainly using advanced models such as deep learning.

[0088] 3. Extract features that affect the model target from the original data. Perform scaling, encoding, and other processing on the features to improve the performance of the model.

[0089] 4. Use the training set data to train the model and obtain the initial parameters of the model.

[0090] 5. Adjust the model parameters, such as learning rate, regularization coefficient, etc., to improve the performance of the model. Use feature selection, feature engineering and other methods to further optimize the model.

[0091] In some embodiments of the present invention, the method further comprises:

[0092] Extracting financial indicator data from the enterprise's financial statements, wherein the financial indicator data includes current assets, current liabilities, total liabilities and total assets;

[0093] Calculate the current ratio based on the current assets and current liabilities;

[0094] The debt-to-asset ratio is calculated based on the total liabilities and total assets.

[0095] In some embodiments of the present invention, in the step of calculating the current ratio based on the current assets and current liabilities, current assets / current liabilities are calculated as the current ratio; total liabilities / total assets are calculated as the debt-to-asset ratio.

[0096] In summary, the present invention combines the company's internal business data to solve the problem of data acquisition; through the company's internal artificial intelligence technology and big data technology, customer service, business and operation and maintenance personnel use real data as the basis to continuously train the model to solve the problem of model updating and maintenance; the present invention establishes a big data model through effective data such as corporate transaction data and repayment records to predict post-loan risks and prevent the generation and accumulation of non-performing assets.

[0097] An embodiment of the present invention also provides a financing data analysis system, which includes a computer device, wherein the computer device includes a processor and a memory, wherein the memory stores computer instructions, and the processor is used to execute the computer instructions stored in the memory. When the computer instructions are executed by the processor, the system implements the steps implemented by the method described above.

[0098] The embodiment of the present invention also provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps implemented by the aforementioned financing data analysis method are implemented. The computer-readable storage medium can be a tangible storage medium, such as a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a floppy disk, a hard disk, a removable storage disk, a CD-ROM, or any other form of storage medium known in the technical field.

[0099] It should be understood by those skilled in the art that the exemplary components, systems and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software or a combination of the two. Whether it is performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present invention are programs or code segments used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link via a data signal carried in a carrier.

[0100] It should be clear that the present invention is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present invention.

[0101] In the present invention, features described and / or illustrated for one embodiment may be used in the same or similar manner in one or more other embodiments, and / or combined with or replace features of other embodiments.

[0102] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the embodiments of the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A financing data analysis method, characterized in that: The steps of the method include: Retrieving original financing data from a database, wherein the original financing data includes numerical data, classification data, and time data; Performing quantitative feature extraction on the numerical data, wherein the numerical data includes financing amount, historical borrowing times, repayment amount and trade data; Extracting qualitative features from the classified data, wherein the classified data includes enterprise industry, enterprise size, and financing purpose; The time data is subjected to time feature extraction, and the financing time data is decomposed during the time feature extraction process.

2. The financing data analysis method according to claim 1, characterized in that: The steps of the method also include: integrating the data obtained through quantitative feature extraction, qualitative feature extraction and qualitative feature extraction, inputting the data obtained through data integration into a pre-trained machine learning model, and outputting the financing risk value through the pre-trained machine learning model.

3. The financing data analysis method according to claim 2, characterized in that: In the step of data integration of the data obtained through quantitative feature extraction, qualitative feature extraction and qualitative feature extraction: In the process of extracting the quantitative features, the mean and standard deviation of the financing amounts in multiple years are calculated.

4. The financing data analysis method according to claim 2, characterized in that: In the step of integrating the data obtained through quantitative feature extraction, qualitative feature extraction and qualitative feature extraction: In the process of extracting qualitative features from the categorical data, label encoding or one-hot encoding is used to extract features from the categorical data.

5. The financing data analysis method according to claim 2, characterized in that: In the step of integrating the data obtained through quantitative feature extraction, qualitative feature extraction and qualitative feature extraction: In the process of extracting time features from the classified data, the financing time data is decomposed into the year, quarter and month.

6. The financing data analysis method according to claim 2, characterized in that: In the step of data integration of the data obtained through quantitative feature extraction, qualitative feature extraction and qualitative feature extraction: In the process of extracting time features from the classified data, the financing frequency is calculated based on the time points of multiple financing time data.

7. The financing data analysis method according to any one of claims 2 to 6, characterized in that: The machine learning model includes a first-stage processing model and a second-stage processing model. The first-stage processing model includes sub-models corresponding to quantitative features, qualitative feature extraction and time features. The data obtained by data integration is input into the pre-trained machine learning model. In the step of outputting the financing risk value through the pre-trained machine learning model, the extracted quantitative features, qualitative feature extraction and time features are respectively input into the pre-trained sub-models. The outputs of multiple sub-models are combined and input into the second-stage processing model. The financing risk value is output through the second-stage processing model.

8. The financing data analysis method according to claim 1, characterized in that: The method also includes the steps of: Extracting financial indicator data from the enterprise's financial statements, wherein the financial indicator data includes current assets, current liabilities, total liabilities and total assets; Calculate the current ratio based on the current assets and current liabilities; The debt-to-asset ratio is calculated based on the total liabilities and total assets.

9. The financing data analysis method according to claim 8, characterized in that: In the step of calculating the current ratio based on the current assets and current liabilities, current assets / current liabilities are calculated as the current ratio; total liabilities / total assets are calculated as the debt-to-asset ratio.

10. A financing data analysis system, characterized in that: The system includes a computer device, which includes a processor and a memory. The memory stores computer instructions. The processor is used to execute the computer instructions stored in the memory. When the computer instructions are executed by the processor, the system implements the steps implemented by the method as described in any one of claims 1 to 9.

Images