Intelligent financial data analysis processing method and device

Abstract

The application discloses an intelligent financial data analysis processing method and device; the scheme comprises the following steps: obtaining target financial report data; positioning layout information of financial key data in the target financial report data, the layout information being used for identifying the same logical region of the position of at least one financial key data in each financial data item; analyzing the target financial report data based on a first analysis frequency to obtain a first financial data item sequence; extracting key financial data items from the first financial data item sequence based on the layout information to obtain a key financial data item sequence; evaluating the correlation degree of each key financial data item and other key financial data items in the key financial data item sequence, and obtaining a target key financial data item from the key financial data item sequence according to the correlation degree; and obtaining financial key data in the target financial report data according to the target key financial data item, so that the efficient and accurate acquisition of the financial key data can be realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of information technology and intelligent financial technology, and particularly relates to an intelligent financial data analysis processing method and device. BACKGROUND

[0002] In the field of enterprise financial management today, with the diversification of business and the continuous expansion of scale, financial statement data becomes increasingly complex and huge. Financial statements contain numerous financial data items, which reflect the financial position of the enterprise from different angles, such as detailed information of assets, liabilities, profits, etc. For enterprise managers, investors and financial analysts, it is crucial to obtain valuable information from these massive financial statement data.

[0003] At present, in the processing of financial statement data, the traditional method mainly relies on manual analysis or simple statistical tools. Manual analysis is inefficient and prone to errors, especially when faced with a large amount of complex data, it is difficult for analysts to fully and accurately grasp the relationship between data. Although simple statistical tools can perform some basic data processing, such as summation and average, they have limited ability for deep data mining and relationship analysis. Some existing data analysis software can perform more complex data analysis, but often lack special design for the characteristics of financial statement data, and cannot accurately locate and effectively extract key financial data when processing key financial data. SUMMARY

[0004] The main purpose of the present application is to provide an intelligent financial data analysis processing method and device, which can realize efficient and accurate acquisition of key financial data.

[0005] To achieve the above purpose, the embodiments of the present application provide an intelligent financial data analysis processing method, which comprises:

[0006] Obtaining target financial statement data, at least one financial data item in a plurality of financial data items of the target financial statement data contains key financial data;

[0007] Locating the layout information of the key financial data in the target financial statement data, the layout information being used to identify the same logical area of the position of at least one key financial data in each financial data item;

[0008] Analyzing the target financial statement data based on a first analysis frequency to obtain a first financial data item sequence;

[0009] Extracting key financial data items from the first financial data item sequence based on the layout information to obtain a key financial data item sequence;

[0010] evaluate a correlation degree of each key financial data item in the sequence of key financial data items with other key financial data items, and acquire a target key financial data item from the sequence of key financial data items according to the correlation degree;

[0011] acquire financial key data in the target financial report data according to the target key financial data item.

[0012] Correspondingly, the embodiment of the application further provides an intelligent financial data analysis processing device, which comprises:

[0013] an acquisition module, configured to acquire target financial report data, wherein at least one financial data item in a plurality of financial data items of the target financial report data contains financial key data;

[0014] a positioning module, configured to position layout information of the financial key data in the target financial report data, wherein the layout information is used to identify a same logical area of a position of the at least one financial key data in each financial data item;

[0015] an analysis module, configured to analyze the target financial report data based on a first analysis frequency to obtain a first sequence of financial data items;

[0016] a first data extraction module, configured to extract key financial data items from the first sequence of financial data items based on the layout information to obtain a sequence of key financial data items;

[0017] an evaluation module, configured to evaluate a correlation degree of each key financial data item in the sequence of key financial data items with other key financial data items, and acquire a target key financial data item from the sequence of key financial data items according to the correlation degree;

[0018] a second data extraction module, configured to acquire financial key data in the target financial report data according to the target key financial data item.

[0019] In summary, the technical solution of this application first obtains the target financial statement data, identifies the financial data items containing key financial data, and locates the layout information of the key financial data; then, it analyzes the statement data at a first analysis frequency to obtain a first financial data item sequence system, systematically organizing the data; and extracts key financial data items from the sequence based on the layout information to form a sequence, which can accurately separate important data and avoid interference from irrelevant data; it assesses the correlation between key financial data items and obtains the target key financial data items accordingly; finally, it obtains the key financial data based on the target key financial data items. This method can efficiently and accurately obtain truly valuable information from massive and complex financial statement data, achieving efficient and accurate acquisition of key financial data. It overcomes the shortcomings of traditional manual analysis, such as low efficiency and susceptibility to errors, as well as the lack of specificity of existing tools, greatly improving the efficiency, accuracy, and depth of financial analysis, and providing a more reliable and valuable basis for corporate decision-making. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a schematic diagram of a scenario for the intelligent financial data analysis and processing method in the embodiments of this application;

[0022] Figure 2 A flowchart of the intelligent financial data analysis and processing method provided in the embodiments of this application;

[0023] Figure 3 A flowchart illustrating the layout information positioning provided in the embodiments of this application;

[0024] Figure 4 A flowchart illustrating the sequence of key financial data items provided in the embodiments of this application;

[0025] Figure 5 A flowchart illustrating the correlation fluctuation situation provided in the embodiments of this application;

[0026] Figure 6 A schematic diagram of the structure of the intelligent financial data analysis and processing device provided in the embodiments of this application;

[0027] Figure 7 A schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0028] With reference to the drawings, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments of the present application. Based on the embodiments in the present application, all the other embodiments obtained by a person of ordinary skill in the art without creative work fall within the scope of the present application.

[0029] The present application provides a kind of intelligent financial data analysis processing method and device, will be described in detail as follows.

[0030] In the present application, the intelligent financial data analysis processing refers to using a series of intelligent technical means, algorithm and tool, to the various data in financial statements in-depth mining, analysis, collation and optimization, to guarantee the integrity and accuracy of data under the premise, quickly, efficiently obtain valuable financial information operation process.This process is mainly aimed at each data part in financial statements, including but not limited to asset data, liability data, profit data, cash flow data etc.

[0031] For example, in the aspect of asset data, the intelligent financial data analysis processing method provided by the present application can first obtain the target financial statement data containing asset related data, wherein data items such as fixed assets, current assets, etc. may contain financial key data. By locating the layout information of these financial key data (such as the net value of fixed assets, the turnover rate of current assets, etc.) in each data item, it is like drawing a map of asset data. Based on a specific analysis frequency, the target financial statement data is analyzed to obtain a data item sequence, and then the key asset data item is extracted from the sequence according to the layout information to form a sequence. For example, if the investment efficiency of enterprise fixed assets is to be analyzed, the correlation between each asset data (such as the purchase cost of different types of fixed assets, the depreciation condition, etc.) in the key asset data item sequence can be evaluated, and the target key asset data item can be obtained according to the correlation, such as those fixed asset data that have greater impact on the overall asset yield. Finally, the financial key data related to the investment efficiency of fixed assets can be accurately obtained according to these target key asset data items, so as to deeply understand the asset condition of the enterprise.

[0032] For example, in the profit data section, the intelligent financial data analysis and processing method provided by the embodiments of the present application can obtain target financial statement data, locate key data layout information in the profit data, such as the logical positions of key data such as main business profit and other business profit in the report. Using a suitable analysis frequency to obtain a data item sequence, and then extracting key profit data items. For example, when analyzing the stability of the profit structure of an enterprise, the correlation between key profit data items is evaluated, such as the correlation between main business profit and other business profit. If the correlation between main business profit and other business profit is low, and the proportion of main business profit is low, it may mean that the profit structure of the enterprise is unstable, and the profit source is too dependent on non-main business. In this way, accurate profit-related financial key data is obtained according to target key profit data items, providing a basis for enterprise profit management.

[0033] For example, in terms of cash flow data, the intelligent financial data analysis and processing method provided by the embodiments of the present application can obtain financial key data in each data item of cash inflow and outflow when obtaining target financial statement data, locate the layout information of these financial key data, and then analyze the cash flow data item sequence with a specific analysis frequency, and then extract key cash flow data items. For example, when analyzing the short-term liquidity of an enterprise, the correlation between key cash flow data items is evaluated, such as the correlation between operating cash flow, investment cash flow, and financing cash flow. If the correlation between operating cash flow and financing cash flow is too high, and the operating cash flow is negative for a long time, it may indicate that the short-term liquidity of the enterprise is at risk. According to the target key cash flow data items, related financial key data is obtained, which helps enterprises to make financial planning and risk prevention in advance.

[0034] As shown in Figure 1 An intelligent financial data analysis and processing scenario is provided, which includes an enterprise financial database, a data preprocessing module, and a financial data relationship analysis device. The data preprocessing module and the financial data relationship analysis device are connected through an enterprise internal network.

[0035] The enterprise financial database stores a large amount of financial statement data, which covers the financial information of each subsidiary and different business departments under the group, including balance sheets, profit statements, cash flow statements, and many other financial statement data. Each data item in these reports is an important basis for analyzing the financial situation of an enterprise, such as fixed asset net value and current asset amount in the balance sheet, main business income and cost in the profit statement, and operating cash inflow and outflow in the cash flow statement.

[0036] A data preprocessing module is configured to perform preliminary processing on raw financial statement data obtained from an enterprise financial database. The module employs advanced data cleaning algorithms to identify and remove erroneous values, duplicate values, and incomplete data from the data. For example, during the entry of some financial statements, erroneous numbers may be entered due to human error, or some data may be entered repeatedly due to system failure. The data preprocessing module can quickly detect and correct these problems. At the same time, it also performs standardization processing on the data, converting data in different formats and different units of measurement into a unified format that is convenient for subsequent analysis. For example, financial data reported by different subsidiaries in different currencies is converted into a unified accounting base currency amount, and financial data recorded in different date formats is converted into a standard date format. The processed data is sent to the intelligent analysis and decision center and the financial data relationship analysis device.

[0037] The financial data relationship analysis device is configured to locate layout information of the financial key data in the target financial statement data, the layout information being used to identify the same logical area where at least one financial key data is located in each financial data item; performing analysis on the target financial statement data based on a first analysis frequency to obtain a first financial data item sequence; extracting key financial data items from the first financial data item sequence based on the layout information to obtain a key financial data item sequence; evaluating the correlation degree of each key financial data item in the key financial data item sequence with other key financial data items, and obtaining a target key financial data item from the key financial data item sequence according to the correlation degree; obtaining the financial key data in the target financial statement data according to the target key financial data item, and outputting and displaying the financial key data, so that financial personnel make decisions based on the financial key data.

[0038] The financial data relationship analysis device can be a computer device, such as a personal computer, a server, etc. It can also be composed of one or more computer devices, or a cloud computing platform, etc.

[0039] Reference Figure 2 , Figure 2 is a flowchart of an intelligent financial data analysis and processing method provided by the embodiments of the present application. The execution subject of the method can be a computer device, which can be a computer device or a cluster composed of multiple computer devices. The computer device can be a terminal device or a server, etc. The intelligent financial data analysis and processing method provided by the embodiments of the present application specifically includes the following steps:

[0040] Step S10: Obtain target financial statement data, at least one financial data item in a plurality of financial data items of the target financial statement data containing financial key data.

[0041] In this application, the target financial statement data is a data set contained in a document prepared by an enterprise in a specific accounting period in accordance with accounting standards, which is used to reflect information such as the enterprise's financial position, operating results, and cash flow. Common target financial statement data comes from major financial statements such as the balance sheet, income statement, and cash flow statement.

[0042] In an embodiment of this application, a financial data item is a basic element that constitutes a financial statement. For example, in a balance sheet, "cash" is a financial data item representing the enterprise's holding of monetary funds; "accounts receivable" is another financial data item referring to the amount of money that the enterprise should collect from the purchasing unit or service-receiving unit for selling goods or providing services.

[0043] In an embodiment of this application, financial key data is data that has a significant impact on the enterprise's financial position, operating results, or cash flow and can reflect the core financial characteristics of the enterprise among numerous financial data items. For example, in assessing the enterprise's debt-paying ability, "asset-liability ratio" (ratio of total liabilities to total assets) is financial key data; in analyzing the enterprise's profitability, "net profit margin" (ratio of net profit to operating income) is financial key data.

[0044] In an embodiment, the target financial statement data is obtained through the enterprise's internal financial information management system (EFIMS) or financial database. EFIMS is a software system used by the enterprise to integrate, store, and manage financial data, and each department enters financial data into it according to the prescribed process. The system collects relevant data from different financial statement modules and then filters out the target financial statement data according to the set standards. The advantage of this approach is the accuracy and completeness of the data, because EFIMS follows strict internal financial data entry specifications. The technical effect is to provide a comprehensive and reliable data basis for subsequent financial analysis, ensuring that the analysis results can truly reflect the enterprise's financial position.

[0045] S20: Locate the layout information of the financial key data in the target financial statement data, which is used to identify the same logical area where at least one financial key data is located in each financial data item.

[0046] In this application, layout information refers to the description of the specific location relationship and logical structure of the financial key data in the target financial statement data structure.

[0047] In financial statements, the arrangement of data items is not random, but follows certain accounting principles and financial logic. This logic divides the statement into different areas, each containing a specific type of financial data item. For example, in the balance sheet, the statement is divided into different logical areas according to factors such as the liquidity of assets and the repayment period of liabilities. The asset section, current assets (such as cash, accounts receivable, inventory, etc.) form a logical area because they have similar properties in terms of enterprise operating capital turnover, etc.; non-current assets (such as fixed assets, intangible assets, etc.) form another logical area. The liability section, current liabilities (such as short-term loans, accounts payable, etc.) and non-current liabilities (such as long-term loans, long-term accounts payable, etc.) also form different logical areas.

[0048] Layout information identifies the location of financial key data in these logical areas. Take the current ratio (current assets / current liabilities) as an example, which measures the short-term solvency of the enterprise. The relevant financial key data items, current assets and current liabilities, have a clear location in the balance sheet, with current assets located in a specific logical area of the asset section (usually at the top of the statement) and current liabilities located in a specific logical area of the liability section (also at the top of the statement). This location relationship in the respective data items is the embodiment of the same logical area identified by the layout information. This helps to accurately find and understand the financial key data related to the current ratio, making it easier to analyze the short-term solvency of the enterprise.

[0049] For example, in the income statement, according to the order of profit calculation, starting from operating income, subtract operating costs, sales expenses, management expenses, financial expenses, etc. to get operating profit, and then adjust through non-operating income and expenses to get total profit, and finally get net profit. For example, the operating profit margin (operating profit / operating income) is a financial key data. The location of operating profit and operating income in the income statement is in a specific logical area relationship. Operating income is located at the top of the income statement, which is the starting data item for calculating profit; operating profit is the data item obtained after a series of cost and expense deductions, and its location is below operating income and has a clear calculation relationship with operating income. This location relationship in the income statement is the embodiment of the layout information for the financial key data of operating profit margin.

[0050] In the embodiments of the present application, the layout information of the financial key data is constructed based on the compilation rules and logic of the financial statements. For example, in the balance sheet, according to the basic equation of "assets = liabilities + owner's equity", the arrangement of data items has a certain order. The asset part usually lists the current assets first, and then the non-current assets; the liability part lists the current liabilities first, and then the non-current liabilities. This arrangement order forms different logical areas, and the financial key data is distributed in these logical areas. For example, for the "current ratio" (ratio of current assets to current liabilities) which measures the short-term solvency of the enterprise, the relative positions of the current assets and the current liabilities in the balance sheet constitute the layout information related to this financial key data.

[0051] The key financial data item is a financial data item related to a specific financial analysis target, used to analyze the financial situation of a certain aspect of the enterprise. For example, when analyzing the operating capital management of the enterprise, the data items related to the "accounts receivable turnover rate" (ratio of operating income to average accounts receivable balance) (operating income, accounts receivable) are key financial data items.

[0052] In an embodiment, a data structure analysis algorithm can be used to locate the layout information. Specifically, first, the storage structure of the target financial statement data is parsed to determine the hierarchical relationship and order between the data items. For example, for a financial statement stored in table form, the algorithm identifies the row, column structure and table header information of the table, thereby determining the position of each data item. Then, according to the pre-defined financial key data type, the corresponding logical area in the parsed structure is determined, and the layout information is obtained. This technical implementation can accurately parse the data structure of financial statements of different formats, ensuring accurate positioning of the layout information. The technical effect is to provide necessary position guidance for subsequent accurate extraction of key financial data items, improving the efficiency and accuracy of data processing.

[0053] Step S30: analyzing the target financial statement data based on the first analysis frequency to obtain a first financial data item sequence.

[0054] In the present application, the first analysis frequency refers to the time interval or period setting for data collection or processing when analyzing the target financial statement data. The determination of this frequency depends on factors such as the operating characteristics of the enterprise, the financial cycle, and the analysis purpose.

[0055] Among them, the financial data item sequence is a set of values of financial data items arranged in a certain order. For example, analyzing the operating income of the enterprise with a monthly first analysis frequency, arranging the operating income values of each month in chronological order forms a financial data item sequence of operating income.

[0056] The first financial data item sequence is a sequence set of each of the plurality of financial data items obtained by analyzing the target financial statement data based on a first analysis frequency. For example, the sequence of the current assets and the sequence of the current liabilities are obtained by analyzing the current assets and the current liabilities in the balance sheet, respectively, according to the first analysis frequency, and the first financial data item sequence is constituted by the sequence of the current assets and the sequence of the current liabilities.

[0057] In an embodiment, the analysis based on the first analysis frequency can be implemented by using a data collection and processing software tool. According to the first analysis frequency set, the tool extracts the values of each of the financial data items from the target financial statement data at a predetermined time point. For example, if the first analysis frequency is quarterly, the tool extracts the values of the relevant financial data items from the balance sheet, the profit sheet and the like at the end of each quarter. Then, the values of each of the financial data items at different time points are arranged in chronological order to form a first financial data item sequence of each of the financial data items. The advantage of this technical implementation is that the data can be systematically obtained and arranged according to a unified frequency, ensuring the time sequence and integrity of the data. The technical effect is to clearly show the trend of the financial data items over time, providing a dynamic data basis for subsequent analysis.

[0058] Step S40: extracting key financial data items from the first financial data item sequence based on the layout information to obtain a key financial data item sequence.

[0059] In the present application, the key financial data item sequence is a sequence composed of key financial data items arranged in a certain order. For example, when analyzing the solvency of an enterprise, the total assets and the total liabilities related to the asset-liability ratio are extracted from the first financial data item sequence as key financial data items, and these key financial data items are arranged in a specific order (such as a time order or an analysis logic order) to obtain a key financial data item sequence related to the solvency analysis.

[0060] In the present application, the key financial data item sequence is a sequence composed of key financial data items arranged in a certain order. For example, when analyzing the solvency of an enterprise, the total assets and the total liabilities related to the asset-liability ratio are extracted from the first financial data item sequence as key financial data items, and these key financial data items are arranged in a specific order (such as a time order or an analysis logic order) to obtain a key financial data item sequence related to the solvency analysis.

[0061] In an embodiment, a rule-based screening algorithm is adopted, in which the screening rules are formulated according to the correspondence between the logical regions determined by the layout information and the first sequence of financial data items. For example, if the layout information indicates that the position index range of a certain key financial data item in the first sequence of financial data items is from the 10th to the 20th data item, then the screening algorithm extracts the corresponding data items from the first sequence of financial data items according to this index range. The extracted multiple key financial data items are arranged in a predetermined order to obtain a sequence of key financial data items. The advantage of this technical implementation is that the key financial data items can be quickly and accurately extracted according to the precise layout information, avoiding the interference of irrelevant data. The technical effect is that the obtained sequence of key financial data items is more compact and targeted, which is conducive to the in-depth analysis of the relationship between the key financial data items.

[0062] Step S50: evaluating the correlation degree of each key financial data item in the sequence of key financial data items with other key financial data items, and obtaining a target key financial data item from the sequence of key financial data items according to the correlation degree.

[0063] In this application, the correlation degree is an index that measures the closeness of the mutual relationship between two or more key financial data items. It reflects the possibility and degree of the influence of the change of one key financial data item on other key financial data items.

[0064] Each key financial data item in the sequence of key financial data items may have different degrees of correlation with other key financial data items. For example, in the sequence of key financial data items of the profit statement, there is a correlation between "sales revenue" and "operating cost". When market demand increases and sales revenue rises, if the cost structure of the enterprise does not change, operating cost will also rise accordingly, and the amplitude and trend of this rise reflect the correlation degree between the two.

[0065] The target key financial data item is a data item that is screened from the sequence of key financial data items according to its correlation degree with other key financial data items, and has more important significance and can better reflect the core financial characteristics of the enterprise.

[0066] In an embodiment, the correlation degree can be evaluated by using a multivariate statistical analysis method. For example, a covariance analysis can be used to calculate the covariance value between the key financial data items, which reflects the degree of linear relationship between two key financial data items. The covariance value is then standardized to obtain a correlation coefficient, which has a value range of -1 to 1. A correlation coefficient close to 1 indicates a strong positive correlation, close to -1 indicates a strong negative correlation, and close to 0 indicates a weak correlation. According to a pre-set correlation degree threshold, for example, setting the threshold to 0.7 (indicating a strong correlation), the key financial data items with a correlation coefficient greater than the threshold are selected as the target key financial data items. This technical implementation is based on rigorous statistical theory and can objectively and accurately evaluate the correlation degree between key financial data items. The technical effect is that by determining the target key financial data items, it can focus on key factors that have a greater impact on the financial status of the enterprise and are closely related, providing more valuable information for further analysis and decision-making.

[0067] Step S60: Obtain the financial key data in the target financial statement data according to the target key financial data items.

[0068] In the implementation of the present application, obtaining the financial key data in the target financial statement data according to the target key financial data items is a process of tracing back and searching. The target key financial data items are key data items selected in the analysis process, which have a specific logical relationship with the original financial data items in the target financial statement data.

[0069] For example, if the target key financial data item is the "return on equity" (ratio of net profit to average net assets), then in the target financial statement data (income statement and balance sheet), the "net profit" and "average net assets" original financial data items need to be obtained. "Net profit" can be obtained from the income statement, and "average net assets" needs to be calculated according to the relevant data in the balance sheet (opening net assets and closing net assets).

[0070] In an embodiment, a mapping relationship between the target key financial data item and the original data item in the target financial statement data can be constructed first. This mapping relationship is established based on financial calculation formulas and logical relationships. For example, for the target key financial data item of "gross profit ratio" ((revenue - cost of goods sold) / revenue), the mapped original data items are "revenue" and "cost of goods sold". Then, according to this mapping relationship, all original data items related to the target key financial data item are queried and obtained in the target financial statement data. The advantage of this technical implementation is that the corresponding original data of the target key financial data item can be accurately found, ensuring the completeness and accuracy of the financial key data. The technical effect is to provide reliable data support for the financial decision-making of the enterprise, because accurate financial key data can reflect the true financial situation and operating results of the enterprise.

[0071] In an embodiment, reference Figure 3 To improve the accuracy of key financial data extraction, the step of "locating the layout information of the financial key data in the target financial statement data" can include:

[0072] Step S201: Analyzing the target financial statement data with a second analysis frequency to obtain a second financial data item sequence of the target financial statement data.

[0073] In this application, the second analysis frequency is a time interval or data processing rhythm for analyzing the target financial statement data, which is different from the first analysis frequency. The setting of the second analysis frequency can depend on various factors, such as different analysis purposes or specific financial decision-making needs. Unlike the first analysis frequency, which may focus on macro trend analysis, the second analysis frequency tends to analyze micro structure or special cases.

[0074] Among them, the financial data item sequence is a set of financial data items arranged in a certain order. In the embodiment of the present application, the second financial data item sequence is obtained by analyzing the target financial statement data with the second analysis frequency. For example, if the first analysis frequency is to analyze the financial data of an enterprise for a year in quarterly units, a macro quarterly data item sequence is obtained. The second analysis frequency may be in weekly units, and is used to analyze a specific business segment or short-term financial fluctuations, so that the second financial data item sequence obtained can more accurately reflect the changes in financial data in the short term.

[0075] In an embodiment, the dynamic data slicing analysis method can be adopted, and specifically, the rules of data slicing can be set according to the second analysis frequency. For example, if the second analysis frequency is in weeks, the tool divides the target financial statement data according to weeks. Then, for each slice, the tool extracts all related financial data items and arranges them in a pre-set order. For example, for balance sheet data, the data items in each slice are arranged in the order of assets, liabilities, and owner's equity; and for profit statement data, the data items are arranged in the order of operating income, cost, and expense. Finally, the data item sequences in each slice are integrated to obtain the second financial data item sequence. The advantage of this technical implementation is that the data can be sliced and arranged flexibly according to different second analysis frequencies, and the sequence reflecting the change of financial data at a specific frequency can be accurately obtained. The technical effect is that more detailed and targeted data information can be provided, which helps to find the characteristics and rules of financial data in short-term or specific situations.

[0076] S202: Obtain a logical mapping diagram of the second financial data item sequence according to a financial data structure detection model.

[0077] In this application, the financial data structure detection model is a model specially used for analyzing the relationship of financial data structure. It is constructed based on a large number of financial data structure patterns and rules.

[0078] In the embodiment of this application, the logical mapping diagram is a graphical or abstract representation of the logical relationship between data items in the second financial data item sequence. The logical mapping diagram can reveal the structural interdependence and data flow of the data items. For example, in a complex enterprise group financial statement, there can be various relationships between the financial data items of different subsidiaries, such as fund transfer and cost allocation. The financial data structure detection model can find these relationships by analyzing the second financial data item sequence and present them in the form of a logical mapping diagram. In the logical mapping diagram, nodes can represent financial data items, and lines and arrows can represent the logical relationship between data items, such as the flow of a cost data item to another profit calculation data item, indicating the influence of the cost on the profit.

[0079] In an embodiment, the financial data structure detection model can be constructed based on a graph algorithm. Specifically, each data item in the second sequence of financial data items can be regarded as a node in a graph; then, by analyzing the calculation logic, accounting rules, and data flow of the financial data, and other factors, the edges between the nodes and the weights of the edges can be determined. For example, if one data item is a direct input for the calculation of another data item, there is an edge between them, and the weight of the edge is determined according to its importance in the calculation. The model constructs a graph structure according to these nodes and edges, and after optimization and visualization processing, the graph structure becomes a logical mapping diagram. The advantage of this technical implementation is that it can accurately capture the complex logical relationships between financial data items and display them in an intuitive graph structure.

[0080] In an embodiment, step S202 can specifically include: inputting the second sequence of financial data items into the financial data structure detection model, performing structural analysis on each data item according to the preset financial logic rules by the model, and obtaining a preliminary analysis result; performing integration and mapping processing on the preliminary analysis result, and generating a logical mapping diagram that can reflect the logical relationships between the financial data items.

[0081] The preset financial logic rules are a series of rules determined according to accounting standards, financial analysis principles, and industry practices. These rules define the reasonable relationships between financial data items, the calculation order, and the specifications of the data structure. For example, in financial statements, the measurement method of assets, the matching principle of cost and income, and other rules are all embodiments of financial logic rules.

[0082] The preliminary analysis result is the preliminary information obtained by the model after analyzing the input second sequence of financial data items according to the financial logic rules. This result contains the preliminary positioning of each data item in the entire financial data structure, the preliminary judgment of the relationship between the data items, and the preliminary evaluation of whether the data items conform to the financial logic rules.

[0083] For example, in the financial data of a manufacturing enterprise, the second sequence of financial data items includes data items such as raw material purchase cost, production equipment depreciation, product sales revenue, etc. The financial logic rules specify that the raw material purchase cost should be associated with the cost accounting in the production process, the production equipment depreciation should be calculated according to a specific depreciation method and related to the production cost, and the product sales revenue should be related to the sales quantity and unit price. When this second sequence of financial data items is input into the financial data structure detection model, the model will analyze each data item according to these rules. For the raw material purchase cost, the preliminary analysis result may determine its position in the cost structure and its potential relationship with other possibly related data items (such as inventory data items); for the production equipment depreciation, the preliminary analysis result will clarify whether its depreciation calculation conforms to the specified method and how it is related to the product production cost; for the product sales revenue, the preliminary analysis result will evaluate whether its relationship with sales-related data items (such as sales channel cost) is reasonable.

[0084] In an embodiment, the preset financial logic rules can be encoded in a form that the rule engine can recognize, for example, written in a rule language (such as the Drools rule language). Then, when the second sequence of financial data items is input into the financial data structure detection model, the rule engine in the model will read each data item in turn and match it with the encoded financial logic rules. For each data item, the rule engine will perform the corresponding structural analysis operation according to the matching result. For example, if a data item is the purchase cost of fixed assets, the rule engine will analyze it according to the financial logic rules of fixed assets (such as the rules for recognition, measurement, capitalization, etc. of purchase cost) to determine its preliminary role and relationship in the financial data structure. Finally, the analysis results of all data items are integrated to obtain the preliminary analysis result. The advantage of this technical implementation is that it can flexibly handle complex financial logic rules and can easily update and maintain the rules.

[0085] In this application, the integration and mapping process is a further processing of the preliminary analysis result. Integration is to summarize and sort the scattered information about each data item in the preliminary analysis result, so that it forms a more orderly overall structure. For example, in the preliminary analysis result, each data item may have been individually analyzed for its partial relationship with other data items, and the integration process is to summarize these relationships to form a more comprehensive relationship network.

[0086] The mapping processing is to present the integrated information in a direct and logical relationship reflecting manner. The result of the presentation is a logical mapping diagram. The logical mapping diagram is an abstract representation that shows the mutual relationship between the financial data items in a graphical or logical structure. For example, in the logical mapping diagram, the financial data items can be represented by nodes, and the logical relationship (such as the causal relationship, the calculation relationship, the dependency relationship, etc.) between the data items can be represented by lines.

[0087] For example, taking a chain retail enterprise as an example, the preliminary analysis result can include the relevant information of the data items such as the sales revenue, the procurement cost, and the operating expense of each store. In the integration process, the same type of data items of different stores are summarized, and the correlation between them is sorted out, such as the relationship between the sales revenue of a store and the procurement cost and operating expense of the store, and the comparison relationship of the sales revenue between different stores. In the mapping processing, these relationships are represented in the form of a logical mapping diagram. For example, the sales revenue node of a store can have a line with the corresponding procurement cost node and operating expense node, indicating the cost-benefit relationship between them; the sales revenue nodes of different stores can also have lines, indicating the sales performance comparison relationship between the stores.

[0088] In an embodiment, the data items and the relationship information between them can be extracted from the preliminary analysis result. Then, the data items are taken as nodes, and the relationships are taken as edges to construct an initial graph structure. In the construction process, different weights or attributes are set for the edges according to the types of the relationships (such as strong correlation, weak correlation, etc.). For example, if there is a direct calculation relationship between two data items, the weight of the edge between them can be set to a higher value; if it is an indirect correlation relationship, the weight can be set to a lower value. Finally, the constructed graph structure is optimized and visualized to generate a logical mapping diagram. For example, the layout of the nodes is adjusted to make the logical relationship more clearly displayed, and necessary annotations or labels are added to the nodes and edges to facilitate better understanding.

[0089] S203: Obtain the layout information of the financial key data in the target financial statement data according to the logical mapping diagram.

[0090] In this application, the layout information of the financial key data refers to the position relationship of the financial key data in the target financial statement data. The embodiment of this application can obtain the layout information through a logical mapping diagram. The logical mapping diagram shows the logical relationship between data items, and the layout information of the financial key data can be derived from these logical relationships. For example, if it is found in the logical mapping diagram that a certain financial key data has a specific logical association path with other data items, the position relationship of the financial key data in the entire target financial statement data can be determined according to this path.

[0091] Suppose that in the financial statement of an enterprise, there is a financial key data that is a specific ratio for measuring the long-term investment income of the enterprise. It is found through the logical mapping diagram that the calculation of this ratio involves the initial investment amount of the long-term investment project, investment income, asset impairment reserve and other data items, and the position and connection relationship of these data items in the logical mapping diagram reflects their layout relationship in the target financial statement data. The initial investment amount may be located in the non-current assets part of the balance sheet, the investment income may be associated with other income items in the profit statement, and the asset impairment reserve may be related to the measurement and adjustment of assets. The logical relationship of these data items in the logical mapping diagram can determine the layout information of the financial key data in the target financial statement data.

[0092] In an embodiment, the path search method can be used to obtain the layout information. Specifically, the nodes related to the financial key data (i.e. the nodes corresponding to the data items related to the calculation or correlation of the financial key data) are determined in the logical mapping diagram. Then, starting from these nodes, other related nodes are found along the edges of the logical relationship through path search algorithms (such as depth-first search or breadth-first search algorithm), and the positions and connection orders of these nodes in the logical mapping diagram are recorded. Finally, according to the position information of these nodes in the logical mapping diagram, the layout information of the financial key data in the target financial statement data is derived. The advantage of using the path search method to obtain the layout information is that it can accurately mine the layout information of the financial key data according to the logical relationship of the logical mapping diagram, improving the accuracy and efficiency of obtaining the layout information.

[0093] In an embodiment, in order to improve the accuracy and efficiency of the key financial data items, reference Figure 4 , step S40 can specifically include:

[0094] S401: Extracting the logical area corresponding to the layout information from each first financial data item in the first financial data item sequence to generate the key financial data item of each first financial data item.

[0095] The key financial data item is a data item that is extracted from the first financial data item sequence and is of key significance to a specific financial analysis. The embodiment of the present application determines the key financial data item by extracting the logical region corresponding to the layout information. For example, when analyzing the long-term solvency of an enterprise, the data items related to non-current assets (such as fixed assets, long-term investments, etc.) and long-term liabilities (such as long-term borrowings, long-term accounts payable, etc.) in the balance sheet are very crucial to the calculation of relevant solvency indicators (such as asset-liability ratio, equity ratio, etc.). The first financial data items corresponding to these logical regions are found in the first financial data item sequence, and then determined as key financial data items.

[0096] Taking a real estate development enterprise as an example, the first financial data item sequence includes data items such as land acquisition cost, construction engineering cost, presale income, long-term borrowings, short-term borrowings, etc. If the current layout information is the logical region identifier related to the long-term capital structure of the enterprise, then in this logical region, the first financial data item of long-term borrowings is one of the key financial data items. Because long-term borrowings plays a key role in analyzing the long-term capital sources, debt structure and long-term solvency of the enterprise, etc.

[0097] In an embodiment, a template of the logical region can be constructed according to the layout information, which defines the characteristics, range or pattern of the data items contained in the specific logical region. For example, if the logical region is a region related to the long-term assets of the enterprise, the template can be defined as a pattern containing data items such as fixed assets, intangible assets, long-term investments, etc. Then, each first financial data item in the first financial data item sequence is traversed and matched with the logical region template. If a certain first financial data item meets the definition of the template, it is determined as a key financial data item. For example, when the first financial data item of fixed assets is traversed, it is found that it meets the definition of the long-term asset logical region template, and it is marked as a key financial data item. The embodiment of the present application can accurately extract key financial data items according to the pre-defined logical region template, with high accuracy and operability. The technical effect is to accurately screen the key financial data items corresponding to the layout information from the first financial data item sequence, laying a foundation for subsequent construction of the key financial data item sequence.

[0098] Step S402: According to the logical order of the key financial data items, the key financial data items of each first financial data item are sorted into a key financial data item sequence.

[0099] In the present application, the logical order of the key financial data items is the order determined according to the purpose of financial analysis, the internal relationship between the financial data items, and the preparation logic of the financial statements. For example, when analyzing the profitability of an enterprise, the operating income may be considered first, followed by the operating cost, and then various expenses, and so on. This order is determined based on the preparation logic of the profit statement and the requirements of profitability analysis.

[0100] The sequence of key financial data items is a sequence arranged by the key financial data items extracted from the first sequence of financial data items in the above logical order. The construction of this sequence facilitates subsequent operations such as systematic analysis and correlation assessment of the key financial data items. For example, when analyzing the operating capacity of an enterprise, the sequence of key financial data items is constructed in the logical order of the data items involved in the key financial indicators such as accounts receivable turnover rate and inventory turnover rate, which facilitates the calculation and analysis of these indicators and thus a comprehensive assessment of the operating capacity of the enterprise.

[0101] Taking a manufacturing enterprise as an example, the key financial data items such as operating income, operating cost, inventory, and accounts receivable are extracted from the first sequence of financial data items. According to the logical order of analyzing the operating efficiency of the enterprise, the operating income and operating cost may be arranged first (as they are the basis for calculating gross profit), followed by inventory (which is related to the connection between production and sales), and finally accounts receivable (which involves the collection of sales). In this way, a sequence of key financial data items related to the analysis of operating efficiency is constructed.

[0102] In an embodiment, a sorting rule can be formulated according to the purpose of financial analysis and the logical relationship between the financial data items. For example, for the analysis of the solvency of an enterprise, the rule may provide that the key financial data items of the liability category (in the order of the expiration time of the debt) are arranged first, and then the key financial data items of the asset category (in the order of the liquidity of the assets). Then, the key financial data items of each first financial data item are sorted according to the formulated sorting rule. For example, when analyzing the short-term solvency of an enterprise, the key financial data items of short-term liabilities such as short-term loans and accounts payable are arranged in the order of the expiration time, and then the key financial data items of liquid assets such as cash and accounts receivable are arranged in the order of the liquidity after the data items of the liability category, thereby constructing a sequence of key financial data items. The advantage of this technical implementation is that the key financial data items can be sorted according to clear rules, ensuring the logicality and rationality of the sequence. The technical effect is that the constructed sequence of key financial data items meets the requirements of financial analysis, facilitating subsequent operations such as correlation assessment, and contributing to the accurate analysis of the financial status of the enterprise.

[0103] In an embodiment, the step of "evaluating the correlation of each key financial data item in the sequence of key financial data items with other key financial data items" can specifically include:

[0104] evaluate the correlation between each of the key financial data items other than the first key financial data item and the previous key financial data item of the key financial data item, to obtain the correlation of each of the key financial data items.

[0105] The correlation is an index for measuring the closeness of the mutual relationship between two financial data items. In the embodiments of the present application, the focus is on the correlation between the key financial data items other than the first key financial data item and the previous key financial data item. The correlation can reflect the causality between the data items, the degree of mutual influence, and the coherence in the financial logic, etc. For example, in a sequence of key financial data items related to enterprise cost control and profit analysis, there can be key financial data items such as raw material procurement cost, production labor cost, manufacturing expense, product sales revenue, sales expense, net profit, etc. Here, the production labor cost is correlated with the raw material procurement cost, because the amount of raw material procurement can affect the production scale, and in turn affect the production labor cost. The manufacturing expense can also be correlated with the production labor cost, for example, when the production scale expands, the manufacturing expense will increase accordingly, and at the same time, it will also affect the allocation of labor cost. The product sales revenue is also correlated with the manufacturing expense, production labor cost, etc., because the production cost will affect the product pricing, and in turn affect the sales revenue. The closeness of the correlation is the correlation to be evaluated.

[0106] In an embodiment, in order to improve the accuracy of the correlation evaluation, an autoregressive model (AR model) in time series analysis can be used to evaluate the correlation. First, the data items in the sequence of key financial data items are regarded as time series data, although the concept of "time" here can be a logical order concept rather than a real time order. For each key financial data item other than the first key financial data item, the previous key financial data item is taken as the independent variable, and the key financial data item is taken as the dependent variable. For example, for the production labor cost key financial data item (assuming it is not the first data item in the sequence), the raw material procurement cost (the previous key financial data item) is taken as the independent variable. Then, through the parameter estimation process of the AR model, the influence coefficient of the independent variable on the dependent variable is estimated, which can be used as a measure of the correlation. For example, if the estimated coefficient is large and statistically significant, it means that the correlation between the production labor cost and the raw material procurement cost is high; if the coefficient is small or not significant, it means that the correlation is low.

[0107] In an embodiment, the intelligent financial data analysis and processing method can further include: based on the correlation of each of the key financial data items, obtaining the correlation fluctuation status of each of the key financial data items

[0108] In this case, the step S50 of "obtaining the target key financial data item from the sequence of key financial data items according to the correlation degree" can specifically include:

[0109] According to the correlation degree and the correlation degree fluctuation condition of each key financial data item, the target key financial data item is obtained from the sequence of key financial data items.

[0110] The correlation degree fluctuation condition is a description of the change of the correlation degree in different periods or different business scenarios. This fluctuation condition can reflect the dynamics of the financial situation of the enterprise and the stability of the relationship between the financial data items. For example, in the development process of an enterprise, with the change of market environment, enterprise strategy adjustment or internal management, the correlation degree between two key financial data items (such as sales expense and sales revenue) that are originally closely related may change. If the enterprise launches new marketing activities, it may increase the sales expense investment, and in the early stage of the activities, the correlation degree between the sales expense and the sales revenue may be enhanced, but with the saturation of the market or the response measures of the competitors, this correlation degree may be weakened, and this change process from enhancement to weakening is the embodiment of the correlation degree fluctuation condition.

[0111] Taking the production cost and product quality of an enterprise as an example, it is assumed that there is a sequence of key financial data items including raw material procurement cost, production equipment maintenance cost, quality detection cost and product pass rate. Under normal circumstances, the raw material procurement cost and the product pass rate may have a certain correlation degree, because high-quality raw materials can improve the product pass rate. However, if the enterprise changes the raw material supplier or improves the production process, the correlation degree may fluctuate.

[0112] In an embodiment, the correlation degree fluctuation condition can be obtained by using the sliding window technique combined with statistical analysis. First, a suitable sliding window size is determined, which depends on the frequency of the data and the accuracy requirement of the analysis. For example, if it is quarterly financial data, the sliding window can be set to 3-5 quarters. Then, in each sliding window, the correlation degree between the key financial data items is recalculated. For example, for the two key financial data items of raw material procurement cost and product pass rate, in the first sliding window (assuming it contains data of the first three quarters), the correlation degree value is calculated using the correlation degree evaluation method mentioned earlier (such as autoregressive model, etc.); in the second sliding window (containing data of the second to fourth quarters), the correlation degree value is calculated again.

[0113] Then, the correlation degree values in these different windows are statistically analyzed to describe the fluctuation condition. The standard deviation, coefficient of variation, and other statistical indicators of the correlation degree can be calculated. The larger the standard deviation, the greater the fluctuation of the correlation degree between different windows; the coefficient of variation can be compared between correlation degrees of different data magnitudes, and more comprehensively reflects the fluctuation condition. For example, if the standard deviation of the correlation degree between the raw material procurement cost and the product pass rate is large, it indicates that the relationship between them is unstable in different time periods, and the fluctuation condition is obvious.

[0114] The embodiments of the present application comprehensively consider the correlation degree and the correlation degree fluctuation condition to determine the target key financial data item. The correlation degree reflects the closeness of the conventional relationship between data items, and the correlation degree fluctuation condition reflects the stability of this relationship. For example, in the financial analysis of an enterprise, a data item with a high correlation degree and a small correlation degree fluctuation may have a continuous and important impact on the core financial condition of the enterprise, and is more likely to be determined as a target key financial data item.

[0115] Taking the sequence of key financial data items related to the enterprise's capital flow as an example, it includes short-term loans, cash reserves, accounts receivable turnover rate, accounts payable turnover rate, and other key financial data items. There may be a certain correlation between short-term loans and cash reserves, and the correlation degree fluctuates less in different market environments, indicating that the enterprise has a relatively stable control over the relationship between short-term loans and cash reserves in terms of capital management, and these two data items have a key impact on the liquidity and debt paying ability of the enterprise's capital, and are more likely to become target key financial data items.

[0116] In an embodiment, a correlation degree threshold and a correlation degree fluctuation threshold can be set. For example, the correlation degree threshold is set to 0.6, indicating that only data items with a correlation degree higher than this value are initially considered; the standard deviation threshold of the correlation degree fluctuation is set to 0.1, indicating that data items with a fluctuation condition within this range are more desirable. Then, for each data item in the sequence of key financial data items, it is determined whether its correlation degree meets the correlation degree threshold and whether the correlation degree fluctuation condition meets the fluctuation threshold. If a key financial data item has a correlation degree higher than the correlation degree threshold and the correlation degree fluctuation condition meets the fluctuation threshold (e.g., the standard deviation is less than the set value), it is determined as a target key financial data item. For example, for the accounts receivable turnover rate, if its correlation degree with sales revenue is 0.7 (higher than the correlation degree threshold of 0.6), and the standard deviation of the correlation degree in different business cycles is 0.08 (less than the fluctuation threshold of 0.1), then the accounts receivable turnover rate can be determined as a target key financial data item.

[0117] In an embodiment, in order to improve the accuracy of the correlation degree fluctuation condition, the correlation degree fluctuation condition is obtained by Figure 5 The following method can be used to obtain the correlation degree fluctuation condition:

[0118] Step S5021: constructing a correlation degree sequence according to the correlation degrees of the key financial data items; the correlation degrees in the correlation degree sequence are arranged according to the logical order of the key financial data items.

[0119] In the present application, the correlation degrees of the key financial data items are evaluated by a specific method, which is an index for measuring the close degree of the mutual relationship between the key financial data items.

[0120] The correlation degree sequence is a sequence formed by arranging the above correlation degrees according to the logical order of the key financial data items. The logical order here is the order determined according to the structure of the financial statements, the purpose of financial analysis, and the inherent financial relationship between the data items. For example, when analyzing the profitability of an enterprise, the analysis may be performed according to the order of the key financial data items such as operating income, operating cost, gross profit margin, sales expense, management expense, and net profit. The corresponding correlation degree sequence also arranges the correlation degrees between the data items in this order.

[0121] The construction of the correlation degree sequence by the embodiments of the present application helps to observe and analyze the continuity and logic of the correlation between the key financial data items as a whole. For example, in the financial analysis of a manufacturing enterprise, the key financial data items may include raw material purchase quantity, production equipment utilization rate, product output, product sales quantity, and sales revenue. If the correlation degrees between the raw material purchase quantity and the production equipment utilization rate, the production equipment utilization rate and the product output, and so on are analyzed in the order of the production process and the financial logic, and these correlation degrees are arranged in sequence to form a correlation degree sequence, this sequence can reflect the transmission of the correlation between the financial data items at each link in the entire business process from raw material purchase to product sales.

[0122] In an embodiment, an index can be established for each key financial data item, which corresponds to the position of the key financial data item in the logical order. For example, for the example of the manufacturing enterprise described above, the index of the raw material purchase quantity is 1, the index of the production equipment utilization rate is 2, and so on. Then, an array or data structure is created to store the correlation degrees, the subscript of the array corresponds to the index. After calculating the correlation degree of each key financial data item and the previous key financial data item, the correlation degree is stored in the corresponding array position. For example, after calculating the correlation degree between the raw material purchase quantity and the production equipment utilization rate, it is stored in the array position with index 2 (corresponding to the production equipment utilization rate). In this way, the correlation degree sequence can be constructed according to the logical order of the key financial data items. The advantage of this technical implementation is simple and intuitive, which can accurately construct the correlation degree sequence according to the logical order, and is convenient for subsequent operation and analysis of the correlation degree sequence. The technical effect is to provide an orderly structure for systematically analyzing the correlation between key financial data items, which helps to understand the internal logic of the enterprise financial situation.

[0123] Step S5022: calculating the correlation degree fluctuation value between each correlation degree in the correlation degree sequence except the first correlation degree and the previous correlation degree of the correlation degree, wherein the correlation degree fluctuation value is used to represent the correlation degree fluctuation condition of the key financial data item.

[0124] The correlation degree fluctuation value is an index used to quantify the degree of change of the correlation degree between adjacent positions in the sequence, which can reflect the stability of the correlation between the key financial data items.

[0125] For example, when analyzing the cost structure and profit relationship of an enterprise, assume that there are three key financial data items: raw material cost, production cost, and profit. First, the correlation degree between the raw material cost and the production cost, and the correlation degree between the production cost and the profit are calculated. If the correlation degree between the raw material cost and the production cost is 0.8, and the correlation degree between the production cost and the profit is 0.6, then the fluctuation value between the two correlation degrees can be calculated. The calculation of the fluctuation value can reflect the change of the correlation relationship from the formation of the cost (from raw materials to production) to the generation of the profit.

[0126] In an embodiment, the following algorithm can be used to calculate the correlation degree fluctuation value.

[0127] Let the current correlation degree be r i (i > 1), the previous correlation degree be r i-1 , and the correlation degree fluctuation value v i = |r i-1 -r i-1 |. This absolute value difference calculation method can intuitively reflect the change amplitude between adjacent correlation degrees.

[0128] In an embodiment, in order to improve the accuracy of the key financial data, the step of "obtaining target key financial data items from the sequence of key financial data items according to the correlation degree and correlation degree fluctuation of each key financial data item" can include:

[0129] If the correlation degree of a key financial data item in the sequence of key financial data items is greater than a first set value and the correlation degree fluctuation value is greater than a second set value, the key financial data item is determined as a target key financial data item.

[0130] The sequence of key financial data items is obtained through the previous steps, and is a sequence arranged in a specific order by key financial data items. The correlation degree reflects the closeness of the mutual relationship between key financial data items, and the correlation degree fluctuation value reflects the stability of the correlation. The first set value and the second set value are threshold values set in advance, which are used to screen key financial data items that meet specific requirements as target key financial data items.

[0131] In an embodiment, the determination of the first set value can be based on statistical analysis of historical financial data of the enterprise or the average correlation degree level of the industry. For example, by statistically analyzing the correlation degree between similar financial data items of multiple enterprises in the same industry, a general correlation degree reference range is obtained, and then a suitable first set value is determined according to the specific circumstances of the enterprise. The determination of the second set value can more consider the requirements of the enterprise on the stability of the relationship between financial data items. If the enterprise wants to find data items with not only close correlation but also large fluctuation (which may imply special operating conditions or potential development opportunities), the second set value will be set relatively high accordingly.

[0132] In an embodiment, in order to improve the flexibility and accuracy of the financial data, the step S60 can include:

[0133] If the number of target key financial data items is one, the financial key data in the target key financial data item is obtained by using a financial formula reverse deduction method, and the financial key data in the target key financial data item is determined as the financial key data in the target financial report data.

[0134] If the number of target key financial data items is greater than one, at least two target key financial data items are integrated to generate an integrated data item, the financial key data in the integrated data item is obtained based on a data correlation rule mining method, and the financial key data in the integrated data item is determined as the financial key data in the target financial report data.

[0135] In the field of financial analysis, various financial indicators are calculated through specific financial formulas. When the number of target key financial data items is one, for example, asset turnover, its calculation formula is operating income divided by average total assets. This indicator reflects the efficiency of enterprise asset operation. Using the reverse deduction method of financial formula to obtain financial key data is to determine the data needed to be obtained from the target financial statement data according to the constituent elements of the formula. For asset turnover, we need to obtain operating income from the income statement and asset-related data from the balance sheet to calculate the average total assets. The basis of this method is that the calculation formula of financial indicators is fixed and clear, which reflects the internal logical relationship between financial data.

[0136] Among them, the reverse deduction method of financial formula is a method in the field of financial analysis to determine the source of the constituent elements of a specific financial indicator according to its calculation formula. For any financial indicator, there is a clear calculation formula that reflects the specific mathematical relationship between several financial data items. For example, when calculating the equity net profit rate (net profit divided by shareholder equity), when the equity net profit rate is the target key financial data item, through the reverse deduction method of financial formula, it can be determined that the net profit and shareholder equity, two financial key data, need to be obtained from the financial statements. This method is based on the fixity and logic of the calculation formula of financial indicators, starting from the final comprehensive financial indicator, tracing back to the most basic financial statement data item, and is a reverse deduction process from the result to the constituent elements, providing an effective way to obtain accurate financial key data, thereby laying a foundation for in-depth financial analysis.

[0137] Taking the sales net profit rate as an example, its formula is net profit divided by operating income. This indicator mainly reflects the proportion of net profit in sales revenue, reflecting the profitability of the enterprise. When the sales net profit rate is the target key financial data item, according to the reverse deduction method of financial formula, it is necessary to accurately obtain the net profit and operating income, two financial key data, from the income statement. This method ensures that the most basic financial statement data is traced back from the target key financial data item, thereby providing accurate data sources for subsequent detailed financial analysis.

[0138] In one embodiment, an index database of financial formulae is established. The index database stores all possible financial indicator formulae that can be the target key financial data item, and the financial statement data sources involved in the formulae. For example, for the quick ratio (quick assets divided by current liabilities), the index database records the indicator name quick ratio, its calculation formula, and the fact that quick assets (usually current assets minus inventory) come from the current assets and inventory data items of the balance sheet, and current liabilities also come from the balance sheet. When the target key financial data item is determined to be the quick ratio, the index database is directly queried to obtain the relevant financial statement data item information, and then the corresponding financial key data is extracted from the target financial statement data according to the index information. This approach has the advantages of high efficiency and accuracy, and does not need to re-analyze the relationship between the financial formula and the financial statement data each time, and the index database can be easily updated and expanded to adapt to changes in new financial indicators or financial statement structures.

[0139] In one embodiment, when the number of target key financial data items is greater than one, these data items often reflect the financial status of the enterprise from different dimensions. For example, the target key financial data items include the current ratio, the total asset return rate, and the net profit growth rate. The current ratio reflects the short-term solvency of the enterprise, the total asset return rate reflects the comprehensive utilization effect of assets, and the net profit growth rate reflects the profit growth of the enterprise. The purpose of integrating them to generate integrated data items is to comprehensively evaluate the financial health of the enterprise. The financial key data in the integrated data items is obtained based on data association rule mining, which is to find the association relationship between these different target key financial data items and the financial statement data.

[0140] Taking the current ratio as an example, it involves current assets and current liabilities, which come from the balance sheet; the calculation of the total asset return rate involves total profits and average total assets, the total profits come from the profit statement, and the average total assets come from the balance sheet; the net profit growth rate is related to net profit, which comes from the profit statement. Through data association rule mining, it can be determined which financial key data needs to be obtained from the balance sheet and the profit statement when integrating these target key financial data items. This mining method is based on data mining technology, and through analyzing a large number of financial data instances, it finds the hidden association relationship between data items.

[0141] In an embodiment, the target key financial data items and their known associations with the financial statement data are taken as input data. For example, the information of the association of the asset-liability ratio with the total amount of liabilities, the total amount of assets, the association of the return on net assets with the net profit, the average net assets, etc. is collated. Then, the FP-Growth algorithm is used to mine deeper association rules to determine all the financial key data required in the integration of these target key financial data items. In the target financial statement data, the financial key data is located and extracted according to the mined association rules. The advantage of this technical implementation is that it can effectively mine the association relationships hidden between the target key financial data items and between them and the financial statement data, and is particularly suitable for handling complex multi-data item integration cases.

[0142] Correspondingly, in order to better implement the above method, the embodiments of the present application further provide an intelligent financial data analysis processing device. As shown in the figure, the intelligent financial data analysis processing device 70 comprises an acquisition module 701, a positioning module 702, an analysis module 703, a first data extraction module 704, an evaluation module 705 and a second data extraction module 706, and the specific implementation is as follows: Figure 6

[0143] The acquisition module 701 is used for acquiring target financial statement data, at least one financial data item in a plurality of financial data items of the target financial statement data containing financial key data;

[0144] The positioning module 702 is used for locating the layout information of the financial key data in the target financial statement data, the layout information being used for identifying the same logical area of the position of at least one financial key data in each financial data item;

[0145] The analysis module 703 is used for analyzing the target financial statement data based on a first analysis frequency to obtain a first financial data item sequence;

[0146] The first data extraction module 704 is used for extracting a key financial data item from the first financial data item sequence based on the layout information to obtain a key financial data item sequence;

[0147] The evaluation module 705 is used for evaluating the association degree of each key financial data item in the key financial data item sequence with other key financial data items, and acquiring a target key financial data item from the key financial data item sequence according to the association degree;

[0148] The second data extraction module 706 is used for acquiring the financial key data in the target financial statement data according to the target key financial data item.

[0149] In an embodiment, the positioning module 702 is specifically used for: ​

[0150] analyzing the target financial statement data by using a second analysis frequency to obtain a second sequence of financial data items of the target financial statement data;

[0151] obtaining a logical mapping diagram of the second sequence of financial data items according to a financial data structure detection model;

[0152] obtaining layout information of financial key data in the target financial statement data according to the logical mapping diagram.

[0153] In an embodiment, the positioning module 702 is specifically configured to: input the second sequence of financial data items into a financial data structure detection model, and perform structural analysis on each data item according to a preset financial logic rule by the model to obtain a preliminary analysis result; and perform integration and mapping processing on the preliminary analysis result to generate a logical mapping diagram capable of reflecting logical relationships of each financial data item.

[0154] In an embodiment, the first data extraction module 704 is configured to: extract a logical region corresponding to the layout information from each first financial data item of the first sequence of financial data items to generate a key financial data item of each first financial data item; and sort the key financial data item of each first financial data item into a key financial data item sequence according to a logical order of the key financial data item.

[0155] In an embodiment, the evaluation module 705 is configured to evaluate a correlation degree between each key financial data item in the key financial data item sequence except for a first key financial data item and a previous key financial data item of the key financial data item to obtain a correlation degree of each key financial data item.

[0156] In an embodiment, the first data extraction module 704 is further configured to: obtain a correlation degree fluctuation condition of each key financial data item based on the correlation degree of each key financial data item; and obtain a target key financial data item from the key financial data item sequence according to the correlation degree and the correlation degree fluctuation condition of each key financial data item.

[0157] In an embodiment, the second data extraction module 706 is configured to:

[0158] If the number of the target key financial data items is one, a financial key data in the target key financial data item is obtained by using a financial formula reverse deduction manner, and the financial key data in the target key financial data item is determined as the financial key data in the target financial statement data.

[0159] If the number of the target key financial data items is more than one, at least two of the target key financial data items are integrated to generate an integrated data item, financial key data in the integrated data item is obtained based on a data association rule mining manner, and the financial key data in the integrated data item is determined as the financial key data in the target financial report data.

[0160] The implementation of each module can refer to the foregoing method embodiments, and details are not described herein. The technical effects of the modules and the device are described with reference to the foregoing method embodiments.

[0161] It should be noted that, in specific implementation, each of the above modules can be combined and integrated in one or more modules, or can be implemented as an independent entity. In addition, the above modules can be implemented in the form of hardware or software function modules. The integrated modules, if implemented in the form of software function modules and sold or used as independent products, can also be stored in a computer readable storage medium. The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0162] As shown in Figure 7 The embodiment of the present application further provides a computer device 80, which is characterized by comprising a processor 801 and a memory 802, wherein the memory 802 stores a computer program, and when the computer program is executed by the processor 801, the processor 801 executes the steps of the method described in any one of the above.

[0163] In several embodiments provided in the present application, it should be understood that the disclosed apparatus and method can be implemented by other manners. For example, the apparatus embodiment described above is only schematic, for example, the division of the unit is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units or components shown or discussed can be indirect coupling or communication connection through some interfaces, apparatuses or units, and can be electrical, mechanical or other forms.

[0164] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected to achieve the purposes of the embodiments of the present application according to actual needs. In addition, the functional units in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present alone, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0165] The integrated unit, if realized in the form of a software functional unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the parts that contribute to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0166] The computer program product or computer program provided by the embodiments of the present application includes computer instructions stored in a computer readable storage medium. The processor of the computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions to make the computer device execute the method provided in the embodiments of the present application.

[0167] The terms "first", "second", etc. in the specification and claims and drawings of the embodiments of the present application are used to distinguish different objects, and are not used to describe a specific order. In addition, the terms "include" and any variations thereof are intended to cover

[0168] not exclusively include. For example, a process, method, device, product or equipment including a series of steps or units is not limited to the listed steps or modules, but optionally further includes steps or modules not listed, or optionally further includes other steps or units inherent to the process, method, device, product or equipment.

[0169] Those skilled in the art can understand that the modules and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized in electronic hardware, computer software or a combination of both. In order to clearly illustrate the interchangeability of hardware and software, the components and steps of each example have been described in the above description in a general manner. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. A person skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0170] The method and related apparatus provided by the embodiments of the present application are described with reference to the method flowchart and / or structural schematic diagram provided by the embodiments of the present application. Each flow and / or block in the method flowchart and / or structural schematic diagram, and the combination of the flows and / or blocks in the flowchart and / or block diagram can be realized by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, a special purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate a device implemented in the flowchart Figure 1 The method and related apparatus provided by the embodiments of the present application are described with reference to the method flowchart and / or structural schematic diagram provided by the embodiments of the present application. Each flow and / or block in the method flowchart and / or structural schematic diagram, and the combination of the flows and / or blocks in the flowchart and / or block diagram can be realized by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, a special purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate a device implemented in the flowchart Figure 1 The method and related apparatus provided by the embodiments of the present application are described with reference to the method flowchart and / or structural schematic diagram provided by the embodiments of the present application. Each flow and / or block in the method flowchart and / or structural schematic diagram, and the combination of the flows and / or blocks in the flowchart and / or block diagram can be realized by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, a special purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate a device implemented in the flowchart Figure 1 The method and related apparatus provided by the embodiments of the present application are described with reference to the method flowchart and / or structural schematic diagram provided by the embodiments of the present application. Each flow and / or block in the method flowchart and / or structural schematic diagram, and the combination of the flows and / or blocks in the flowchart and / or block diagram can be realized by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, a special purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate a device implemented in the flowchart Figure 1 The method and related apparatus provided by the embodiments of the present application are described with reference to the method flowchart and / or structural schematic diagram provided by the embodiments of the present application. Each flow and / or block in the method flowchart and / or structural schematic diagram, and the combination of the flows and / or blocks in the flowchart and / or block diagram can be realized by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, a special purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate a device implemented in the flowchart Figure 1 The method and related apparatus provided by the embodiments of the present application are described with reference to the method flowchart and / or structural schematic diagram provided by the embodiments of the present application. Each flow and / or block in the method flowchart and / or structural schematic diagram, and the combination of the flows and / or blocks in the flowchart and / or block diagram can be realized by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, a special purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate a device implemented in the flowchart

[0171] The above disclosure is only the preferred embodiments of the present application, and of course cannot limit the scope of the present application. Therefore, equivalent changes made in accordance with the claims of the present application are still within the scope of the present application.

Claims

1. A method for intelligent financial data analysis processing, characterized in that, The method comprises the following steps: acquiring target financial statement data, at least one of a plurality of financial data items of the target financial statement data containing financial key data; analyzing the target financial statement data by using a second analysis frequency to obtain a second financial data item sequence of the target financial statement data; inputting the second financial data item sequence into a financial data structure detection model, and performing structure analysis on each data item according to a preset financial logic rule by the model to obtain a preliminary analysis result; performing integration and mapping processing on the preliminary analysis result to generate a logic mapping diagram capable of reflecting the logical relationship of each financial data item; acquiring layout information of the financial key data in the target financial statement data according to the logic mapping diagram, the layout information being used to identify the same logical area of the position of at least one financial key data in each financial data item; analyzing the target financial statement data based on a first analysis frequency to obtain a first financial data item sequence; extracting key financial data items from the first financial data item sequence based on the layout information to obtain a key financial data item sequence; evaluating the correlation degree of each key financial data item and other key financial data items in the key financial data item sequence, and acquiring a target key financial data item from the key financial data item sequence according to the correlation degree; if the number of the target key financial data items is one, acquiring the financial key data in the target key financial data item by using a financial formula reverse deduction method, and determining the financial key data in the target key financial data item as the financial key data in the target financial statement data; if the number of the target key financial data items is greater than one, integrating at least two target key financial data items to generate an integrated data item, acquiring the financial key data in the integrated data item based on a data correlation rule mining method, and determining the financial key data in the integrated data item as the financial key data in the target financial statement data.

2. The intelligent financial data analysis processing method of claim 1, wherein, The method further comprises the following steps: extracting the logical area corresponding to the layout information from each first financial data item of the first financial data item sequence to generate the key financial data item of each first financial data item; sorting the key financial data items of each first financial data item into a key financial data item sequence according to the logical order of the key financial data items.

3. The intelligent financial data analysis processing method of claim 1, wherein, The method further comprises the following steps: evaluating the correlation degree between each key financial data item and the previous key financial data item of the key financial data item in the key financial data item sequence except the first key financial data item to obtain the correlation degree of each key financial data item.

4. The intelligent financial data analysis processing method of claim 3, wherein, The method further comprises the following steps: acquiring the correlation degree fluctuation status of each key financial data item based on the correlation degree of each key financial data item. The method further comprises the following steps: The target key financial data item is obtained from the sequence of key financial data items according to the correlation degree and the correlation degree fluctuation condition of each key financial data item.

5. The intelligent financial data analysis processing method of claim 4, wherein, The correlation degree fluctuation condition of each key financial data item is obtained based on the correlation degree of each key financial data item, and the correlation degree fluctuation condition of each key financial data item comprises: A correlation degree sequence is constructed according to the correlation degree of each key financial data item, and the correlation degrees in the correlation degree sequence are arranged in a logical order of each key financial data item; A correlation degree fluctuation value between each correlation degree in the correlation degree sequence except the first correlation degree and the previous correlation degree of the correlation degree is calculated, wherein the correlation degree fluctuation value is used to represent the correlation degree fluctuation condition of the key financial data item.

6. The intelligent financial data analysis processing method of claim 4, wherein, The target key financial data item is obtained from the sequence of key financial data items according to the correlation degree and the correlation degree fluctuation condition of each key financial data item, and the target key financial data item comprises: If the correlation degree of a key financial data item in the sequence of key financial data items is greater than a first set value and the correlation degree fluctuation value is greater than a second set value, the key financial data item is determined as the target key financial data item.

7. An intelligent financial data analysis processing apparatus, characterized by comprising: The device comprises: The obtaining module is configured to obtain target financial statement data, wherein at least one financial data item in a plurality of financial data items of the target financial statement data contains financial key data. The positioning module is configured to analyze the target financial statement data by using a second analysis frequency to obtain a second sequence of financial data items of the target financial statement data; input the second sequence of financial data items into a financial data structure detection model, and perform structure analysis on each data item according to a preset financial logic rule by using the model to obtain a preliminary analysis result; perform integration and mapping processing on the preliminary analysis result to generate a logical mapping diagram capable of reflecting the logical relationship of each financial data item; and obtain layout information of the financial key data in the target financial statement data according to the logical mapping diagram, wherein the layout information is used to identify the same logical area of the position of at least one financial key data in each financial data item. The analysis module is configured to analyze the target financial statement data based on a first analysis frequency to obtain a first sequence of financial data items. The first data extraction module is configured to extract key financial data items from the first sequence of financial data items based on the layout information to obtain a sequence of key financial data items. The evaluation module is configured to evaluate the correlation degree of each key financial data item in the sequence of key financial data items with other key financial data items, and obtain a target key financial data item from the sequence of key financial data items according to the correlation degree. The second data extraction module is configured to: if the number of the target key financial data items is one, acquire financial key data in the target key financial data item by using a financial formula reverse deduction manner, and determine the financial key data in the target key financial data item as financial key data in the target financial report data; if the number of the target key financial data items is more than one, integrate at least two target key financial data items to generate an integrated data item, acquire financial key data in the integrated data item based on a data correlation rule mining manner, and determine the financial key data in the integrated data item as the financial key data in the target financial report data.

Images