Financial analysis system, its program and apparatus

Abstract

This system provides a financial analysis tool that enables the rapid generation of sophisticated financial analysis reports from rough trial balance data, regardless of the type of accounting software or how account names are assigned. [Solution] The system has a computer perform the following steps: a first step ST12 to obtain local account classifications and local accounts from trial balance data; a second step ST13 to identify standard account classifications based on the first correspondence table; a third step ST14 to identify standard accounts through a multi-matrix determination combining name keywords and priority; a fourth step ST15 to associate the identified standard accounts with the items in the report; and a fifth step ST16 to generate the report by assigning the values ​​of local accounts to the report based on the association. In addition, the system records and learns user changes to the linking settings and automatically applies them from the next time onward.

Description

[Technical Field]

[0001] The present invention relates to a financial analysis system, a program therefor and an apparatus therefor that enable rapid generation of advanced financial analysis reports from coarse trial balance data, regardless of the type of accounting software and the naming convention of account items. [Background Art]

[0002] In conventional financial analysis, the format of trial balances output by each accounting software used by a company varies widely, and furthermore, accounting staff of each company assign unique account names such as "Futsu Mitsubishi" or "Payment fee transfer" based on their own judgment, so a great deal of manual processing was required when creating standard analysis reports. This process is extremely personnel-dependent and time-consuming, and has been a major factor preventing managers from quickly grasping numerical values and making judgments. [Prior Art Documents] [Patent Documents]

[0003] [Patent Document 1] Japanese Unexamined Patent Publication No. 2024-88573 [Summary of the Invention] [Problem to be Solved by the Invention]

[0004] The present invention has been made in view of the above-mentioned problems of the prior art, and an object of the present invention is to provide a financial analysis system that can autonomously identify an accounting structure from heterogeneous trial balance data, absorb notation variations and specification differences among accounting software, and quickly create financial analysis reports that conform to the actual operating conditions of a business. [Means for Solving the Problem]

[0005] The present invention provides a financial analysis system in which a computer performs the following steps: a first step of obtaining local account classifications and local accounts belonging to said local account classifications within trial balance data; a second step of identifying standard account classifications corresponding to said local account classifications; a third step of identifying standard accounts defined in the report to be generated that correspond to said local accounts; a fourth step of associating said standard accounts with accounts in the report; and a fifth step of generating the report by assigning the values ​​of said local accounts to accounts in the report based on the association in the fourth step. With this configuration, the first and second steps absorb the differences in the types of accounting software used to create trial balances, and the third step absorbs the differences in how users utilize the accounting software (how accounts are assigned), thereby enabling the generation of appropriate financial analysis data from trial balance data that is otherwise incomplete.

[0006] Preferably, the first step involves identifying the hierarchical structure of the local account classification and, based on the identified hierarchical structure, identifying the local accounts belonging to the local account classification. This configuration allows for the identification of the hierarchical structure of trial balance data, enabling accurate extraction of parent-child relationships—which accounts belong to which categories (parents)—even from flat data without master information.

[0007] Preferably, the local account classification to which the local account belongs is identified based on the name or location of the total column in the hierarchical structure to which the local account applies. This configuration allows for the autonomous reconstruction of the accounting hierarchical structure by analyzing the names of the total columns and the data placement within the trial balance using a proprietary rule engine.

[0008] Preferably, the second step identifies the standard account classification corresponding to the local account classification based on a first correspondence table obtained in advance that shows the correspondence between the local account classification and the standard account classification. With this configuration, by using the first correspondence table, classification names uniquely defined by the accounting software can be accurately linked to the system's standard account classifications, completely absorbing differences in specifications between software programs.

[0009] Preferably, the third step identifies the standard account corresponding to the local account based on some characters (name keywords) within the local account. This structure allows for the identification of standard accounts based on their essential meaning, even for accounts with variations in spelling, by extracting specific keywords such as "ordinary" or "current" from account names.

[0010] Preferably, the third step identifies the standard account corresponding to the local account based on the priority of the certain characters. This configuration allows for prioritizing keywords, enabling even complex subject names containing multiple keywords to be uniquely linked to the most appropriate standard subject.

[0011] Preferably, the third step identifies the standard account corresponding to the local account based on a second correspondence table obtained in advance that shows the correspondence between the local account and the standard account. This configuration allows for the rapid and reliable identification of standard subjects based on existing correspondences by prioritizing the referencing of the second correspondence table.

[0012] Preferably, in the fourth step, the computer displays the item names of the forms that correspond to the subject names of the standard subjects, displays a screen for the user to change the item names or linking settings of the forms, and performs a sixth step in which the computer changes the item names of the forms according to the user's instructions entered on the screen. This configuration provides a bridge UI that allows users to redefine (reinterpret) account attributes, such as changing "accounts payable" to "accounts payable," to match their actual business operations. Furthermore, the system records and learns these instructions, ensuring long-term data continuity even if accounting software changes.

[0013] Preferably, the local account classification and local account are defined for each accounting software, and when the accounting software is changed between the previous period and the current period, if the local account classification and local account match between the previous period and the current period, the computer automatically links them, and if the local account classification matches but the local account differs, the computer performs a seventh step in which it displays a screen for the user to manually link them. [Effects of the Invention]

[0014] According to the present invention, differences in accounting software and inconsistencies in account names can be automatically absorbed, and advanced financial analysis materials such as balance sheets by fund and future accounting charts can be generated from low-resolution trial balance data using consistent standards without human intervention. Furthermore, by learning from the user's past decisions, it is possible to improve the accuracy of the analysis while maintaining data continuity. [Brief explanation of the drawing]

[0015] [Figure 1] This is a block diagram showing the hardware configuration of a financial analysis system 10 according to an embodiment of the present invention. [Figure 2] This is a flowchart illustrating the overall processing of the financial analysis system 10 according to an embodiment of the present invention. [Figure 3] Figure 2 is a flowchart illustrating the details of the standard subject classification identification process (step ST13). [Figure 4] Figure 4 is a flowchart illustrating the user-assisted correction (bridge UI) flow. [Figure 5] This figure shows the transition of the correspondence between local account classification, local accounts, standard account classification, and standard accounts. [Figure 6] This figure shows examples of assigning multifaceted attribute labels to standard subjects and manually modifying (redefining) them. [Figure 7]FIG. 13 is a diagram for explaining a process of reflecting values in trial balance data to each form based on the determined setting of the account item master. [Figure 8] FIG. 14 is a diagram showing an example of linking account item masters by user-intervening "data continuity maintenance (bridge UI)" logic. DETAILED DESCRIPTION OF EMBODIMENTS

[0016] Hereinafter, a financial analysis system 10 according to an embodiment of the present invention will be described in detail. The present system autonomously restores an accounting structure from coarse, flat trial balance data, and normalizes the data to a unique standard account system (including accounts defined by the accounting office that prepares forms) with high accuracy, thereby collectively generating multifaceted financial forms. This enables visualization of the essential actual state of business management without being constrained by formal account names on account books.

[0017] <FUNCTION DESCRIPTION OF COMPONENTS> FIG. 1 is a functional block diagram of the financial analysis system 10 according to the embodiment of the present invention. As shown in FIG. 1, the financial analysis system 10 includes a communication unit 75, an input / output unit 77, a memory 79, and a processing unit 81.

[0018] The input / output unit 77 is a terminal or operation unit that receives input of data from the outside. In the present embodiment, trial balance data is input through the input / output unit 77. Further, an input / output unit 78 outputs (displays) the generated form data. The memory 79 stores a program PRG executed by the processing unit 81, and stores data input from the input / output unit 77. The processing unit 81 is a CPU that executes the program PRG stored in the memory 79 to perform processing of the financial analysis system 10 defined in the present embodiment. The processing unit 81 performs the processing defined in the present embodiment based on data stored in the memory 79, and stores the processing result in the memory 79.

[0019] The processing unit 81 refers to the first correspondence table 31 and the second correspondence table 32 stored in the memory 79 and generates standard data for financial analysis in bulk from the heterogeneous trial balance data taken in from the input / output unit 77. Specifically, in the first step, the processing unit 81 analyzes the names and positions of the total columns in the trial balance data using its own rule engine, autonomously reconstructs the hierarchical structure of parent-child relationships between accounts, and extracts local account information.

[0020] In the second step, the processing unit 81 performs a process to identify the standard account classification that corresponds to the local account classification in the trial balance data entered in the first step, based on the first correspondence table. The first correspondence table is pre-acquired data showing the correspondence between local account classifications and standard account classifications, and is created for each accounting software program.

[0021] In the third step, the processing unit 81 identifies the standard account specified in the report to be generated, which corresponds to the local account. At this time, the processing unit 81 performs a process to identify the standard account corresponding to the local account by performing a multi-matrix determination that combines the local account classification, the local account name keyword (including partial matches), and priority (the order in which characters are given priority). Instead of simply matching names, we employ a "data normalization and mapping" logic that infers the nature of subjects from their hierarchical structure (context). This software accurately links user-defined local accounts to standard accounts. Local account classifications and local accounts are defined or set by the system or user for each accounting software.

[0022] In the fourth step, the processing unit 81 associates the standard accounts identified in the third step with the accounts in the report. Attributes such as cash flow are then assigned to the standard accounts in a batch.

[0023] In the fifth step, the processing unit 81 generates a report by assigning the values ​​of the local accounts to the accounts of the report based on the mapping in the fourth step. That is, it integrates the finalized mapping and numerical values ​​and outputs advanced analytical reports such as a balance sheet by fund.

[0024] The following provides a detailed explanation with specific examples. In the first step performed by the processing unit 81, local account classifications and local accounts used by individual companies are obtained from trial balance data in CSV format or the like, which is input from an external source via the input / output unit 77. In this process, the processing unit 81 analyzes the names of the total columns in the trial balance, such as "Total Current Assets" and "Total Cash and Deposits," as well as the indentation and placement of the data, using its own rule engine. As a result, the processing unit 81 autonomously identifies and restores the accounting hierarchical structure (parent-child relationships) of which accounts belong to which categories, from flat data where account master data is missing.

[0025] In the second step performed by the processing unit 81, the first correspondence table 31, which is stored in memory 79 in advance, is referenced to identify the standard account classification that corresponds to the local account classification identified by the analysis. For example, by mapping the unique name "Other Current Assets" used in accounting software to a common drawer called "Other Current Assets" in this system, differences in specifications between software programs are completely absorbed.

[0026] In the third step performed by the processing unit 81, a multi-matrix determination is performed that combines standard account classification information, local account name keywords, and priority to identify system-specific standard accounts corresponding to local accounts. Specifically, the processing unit 81 extracts specific keywords such as "ordinary" and "current" from the account names.

[0027] If the name is "Mitsubishi UFJ Ordinary Deposit," the processing unit 81 detects the keyword "ordinary," compares it with contextual information indicating that it belongs to the standard account classification "cash and deposits," and determines that it is the standard account "ordinary deposit" according to priority. By using hierarchical information—which classification category an item belongs to—rather than simply matching names, extremely high identification accuracy is achieved.

[0028] In the fourth step performed by the processing unit 81, the aggregate rows (attribute labels) in multiple reports, such as the cash flow statement and the balance sheet by fund, are mapped to the determined standard accounts in a batch process. At this time, the processing unit 81 displays a bridge UI screen via the display 51 that presents the standard subject name and the corresponding report item name (attribute) to the user and accepts change instructions. For example, if an account that is treated as "accounts payable" in accounting is actually "accounts payable (accounts purchased)" in business practice, when the user changes the attribute to "accounts payable" on the display 51, the processing unit 81 records the instruction in the memory 79 and learns to automatically apply it during the next import.

[0029] In the fifth step performed by the processing unit 81, based on the identified correspondences and balance figures for each account, financial statements (B / S, P / L), cash flow statements, balance sheets by fund, future accounting charts, the president's performance report, and rating scores are created in a batch and distributed to the display 51 and user terminal devices 21. The processing unit 81 does not simply transcribe numerical data, but performs advanced aggregation processing that reflects the actual business situation based on attribute redefinition, transforming formal numerical data on ledgers into living numerical data that directly impacts business decisions.

[0030] <Overall Processing Flow> Figure 1 is a flowchart illustrating the overall processing of the financial analysis system 10. Step ST11: The processing unit 81 obtains trial balance data, which is output from accounting software as specified by the user, from the input / output unit 77.

[0031] Step ST12: (First step) The processing unit 81 analyzes the string pattern of the acquired data, identifies the local hierarchical structure, and identifies the local account classification and local account.

[0032] Step ST13: (Second step) The processing unit 81 uses the first correspondence table 31 to identify the standard account classification that corresponds to the local account classification.

[0033] Step ST14: (Third step) The processing unit 81 identifies the corresponding standard account for each local account by performing a multi-matrix determination or the like. The processing unit 81 identifies the standard account corresponding to the local account based on some characters (name keywords) within the local account. At this time, the processing unit 81 identifies the standard account corresponding to the local account based on the priority order of some characters within the local account. It's not just a matter of matching names; it allows for mapping that reflects the nature of the subject based on its hierarchical structure (context).

[0034] Step ST15: (Fourth step) The processing unit 81 associates the standard accounts identified in the third step with the accounts in the report. Attributes such as cash flow are then assigned to the standard accounts in a batch.

[0035] Step ST16: (Fifth step) Based on the mapping in the fourth step, the processing unit 81 assigns the values ​​of the local accounts (the account names of the standard accounts) to the accounts in the report and generates the report. In other words, it integrates the finalized mapping and numerical values ​​and outputs advanced analytical reports such as a balance sheet by fund. These documents include financial statements (settlement reports), balance sheets (B / S), income statements (P / L), cash flow statements, balance sheets by funding source, future accounting charts (variable cost income statements), and credit ratings.

[0036] Figure 3 is a flowchart illustrating step ST14 shown in Figure 2. In step ST14, the processing unit 81 identifies the corresponding standard account for each local account through multi-matrix determination, etc. Step ST21: The processing unit 81 analyzes some of the characters (name keywords) within the local account.

[0037] Step ST22: The processing unit 81 obtains the priority order.

[0038] Step ST23: The processing unit 81 identifies a standard account corresponding to the local account based on some characters identified in step ST21, the priority order obtained in step ST22, and the standard account classification identified in step ST13. In other words, it extracts keywords from names and identifies partial matches. From account names, it extracts specific keywords that evoke the nature of the account, such as "ordinary" or "current." Furthermore, if multiple keywords apply, the system determines which element will be the priority for evaluation according to a pre-set "priority order." By combining three axes—contextual information ("standard subject classification"), keywords in the subject name, and priority—the system autonomously infers the nature of the subject.

[0039] The processing unit 81 executes the process in step ST14 as follows, for example. Refer to the Subject Attribute Initial Settings Table: Refer to the initial settings table (correspondence table) that defines "which report and which summary row the attribute will be reflected in" for each standard subject. Batch automatic setting of attribute labels for analysis: Assigns multiple labels necessary for generating cash flow statements, balance sheets by fund, future accounting charts (variable cost / income and fixed cost classifications), etc., in a batch. Correspondence with each report summary row: Create a system where the numerical value of one account is automatically distributed to the appropriate summary row of multiple analysis reports with different characteristics.

[0040] Finalizing the aggregation logic: Based on the assigned attribute labels, the calculation rules for cash flow increases and decreases, and for reclassification based on the nature of the funds, are finalized. Numerical data reflection / import into designated items: The "beginning balance," "debit / credit amounts," and "end-of-month balance" from the trial balance data are imported into the designated locations of the defined report items. Creation of financial statements and credit rating scores: Rather than simply transcribing numbers, we generate monthly financial statements, credit rating scores, and other documents in bulk based on our unique analytical logic.

[0041] Figure 4 is a flowchart illustrating the user-assisted correction (bridge UI) flow. This is a process where humans make fine adjustments to the results of automated assessments to match the actual business situation. Step ST31: Check for account inconsistencies. Check for inconsistencies / need for correction: The accounting staff will verify whether the displayed attribute settings (e.g., "Accounts Payable" attribute) reflect the actual business situation.

[0042] Step ST32: The screen is displayed via the bridge UI based on the accountant's operation. When changes are made to the accounting software, new departments are established, or inconsistencies that make automatic system linking difficult are detected, the processing unit 81 displays a dedicated operation screen.

[0043] Step ST33 On the modification screen, you can redefine / manually correct account attributes. Even if an item is listed as "Accounts Payable" in the ledger, if it is actually a trade payable, the accountant will manually change (redefine) the attribute to "Accounts Payable".

[0044] Step ST34: The processing unit 81 records and learns the instructions. The processing unit 81 records and saves the linking and correction instructions made by the accounting staff and automatically applies them during subsequent data imports, thereby maintaining the long-term continuity of the data.

[0045] The following are specific examples. As shown in Figure 5(A), the processing unit 81 creates an account master from the trial balance data (step ST12). Next, the processing unit 81 sets the standard subject classification from the subject classification comparison table, as shown in Figure 5(B) (step ST13).

[0046] Next, the processing unit 81, based on the priority within the standard account classification, links the standard accounts by partial priority of the account name, as shown in Figure 5(C) (step ST14). Next, as shown in Figure 6(D), the processing unit 81 associates the standard accounts linked in step ST14 with the accounts (items) of the form (step ST15). Furthermore, if a client processes an item as "accounts payable" in their accounting software, but the accounting firm's staff determines that it should be "accounts payable," they will manually change the account setting to "accounts payable."

[0047] Next, as shown in Figures 6(E) and 7(F), the processing unit 81 reflects the values ​​of the trial balance data in each report based on the mapping in step ST15 (the account master of this system) (step ST16).

[0048] As shown in Figure 8, when the accounting software is changed between the previous and current periods, the processing unit 81 automatically links the local account classifications and local accounts if they match between the previous and current periods, and displays a screen for the user to manually link them if the local account classifications match but the local accounts differ. In this case, if the local account classifications do not match, automatic linking may not be performed, and instructions from the user may be required.

[0049] <Effects of the Embodiment> According to the financial analysis system 10, because it autonomously analyzes the strings and structure of trial balance data, advanced financial analysis can be started immediately with data output from any accounting software without the need for complex prior mapping settings.

[0050] According to the financial analysis system 10, the inference of accounts, which is intuitively performed by users of accounting software, can be systematized through multi-matrix determination, and the system can be made more resilient to variations in the notation of account names.

[0051] According to the financial analysis system 10, the redefinition learning function via the bridge UI allows for the continuous output of actual-fact figures that contribute to correct management decisions, even when there is a discrepancy between accounting procedures and actual business operations.

[0052] According to the financial analysis system 10, by linking to a single standard account, multiple financial reports of different natures (such as future accounting and cash flow analysis) can be generated in a consistent manner, dramatically improving the efficiency of management analysis and enabling rapid decision-making.

[0053] The present invention is not limited to the embodiments described above. In other words, those skilled in the art may make various modifications, combinations, subcombinations, and substitutions with respect to the components of the embodiments described above, within the technical scope of the present invention or its equivalents.

[0054] For example, the processing unit 81 of the financial analysis system 10 may collaborate with a large external account database to complement the accuracy of the multi-matrix determination and perform automatic linking based on similarity determination. Furthermore, the financial analysis system 10 may be provided as a cloud-based service, and the processing unit 81 may be configured as a distributed computing environment consisting of multiple physical servers.

[0055] Furthermore, the financial analysis system 10 may be equipped with an AI advisor function in which the processing unit 81 automatically proposes measures to improve cash flow and the possibility of obtaining loans based on the generated rating score. Furthermore, in the third step, the processing unit 81 of the financial analysis system 10 may preferentially refer to the second correspondence table 32, which shows the direct correspondence between local accounts and standard accounts obtained in advance, to identify the accounts. [Explanation of symbols]

[0056] 10…Financial analysis system 75... Communications Department, 77…Input / output section 79…Memory 81… Processing Section

Claims

1. The first step is to obtain the local account classification and the local accounts belonging to that local account classification within the trial balance data, A second step is to identify a standard account classification corresponding to the local account classification, A third step involves identifying the standard account specified in the report to be generated, which corresponds to the aforementioned local account; A fourth step involves associating the standard accounts identified in the third step with the accounts in the forms, Based on the correspondence in the fourth step, the fifth step involves assigning the values ​​of the local accounts to the accounts in the report and generating the report. A computer-based financial analysis system.

2. The second step described above is: Identify the hierarchical structure of the aforementioned local account classification, Based on the identified hierarchical structure, identify the local accounts belonging to the local account classification. The financial analysis system according to claim 1.

3. Based on the name or location of the total column to which the local account belongs within the aforementioned hierarchical structure, the local account classification to which the local account belongs is identified. The financial analysis system according to claim 2.

4. The second step involves identifying the standard account classification corresponding to the local account classification based on a first correspondence table obtained in advance that shows the correspondence between the local account classification and the standard account classification. The financial analysis system according to claim 1.

5. The third step involves identifying the standard account corresponding to the local account based on some of the characters within the local account. The financial analysis system according to claim 1.

6. The third step described above is: Based on the priority order of some of the aforementioned characters, the standard account corresponding to the local account is identified. The financial analysis system according to claim 5.

7. The third step described above is: Based on a second correspondence table obtained in advance that shows the correspondence between the local account and the standard account, the standard account corresponding to the local account is identified. The financial analysis system according to claim 1.

8. A sixth step involves displaying the item names of the forms associated with the subject names of the standard subjects in the fourth step, displaying a screen for the user to change the item names or linking settings of the forms, and changing the item names of the forms according to the user's instructions entered on the screen. The financial analysis system according to claim 1, wherein the computer performs the following.

9. The aforementioned local account classifications and local accounts are defined for each accounting software program. Step 7: If the accounting software is changed between the previous and current periods, if the local account classification and local account match between the previous and current periods, they are automatically linked. If the local account classification matches but the local account differs, a screen is displayed for the user to manually link them. The financial analysis system according to claim 1, wherein the computer performs the following.

10. The first step is to obtain the local account classification and the local accounts belonging to that local account classification within the trial balance data, A second step is to identify a standard account classification corresponding to the local account classification, A third step involves identifying the standard account specified in the report to be generated, which corresponds to the aforementioned local account; A fourth step involves associating the standard accounts identified in the third step with the accounts in the forms, Based on the correspondence in the fourth step, the fifth step involves assigning the values ​​of the local accounts to the accounts in the report and generating the report. A method of financial analysis performed by a computer.

11. The first step is to obtain the local account classification and the local accounts belonging to that local account classification within the trial balance data, A second step is to identify a standard account classification corresponding to the local account classification, A third step involves identifying the standard account specified in the report to be generated, which corresponds to the aforementioned local account; A fourth step involves associating the standard accounts identified in the third step with the accounts in the forms, Based on the correspondence in the fourth step, the fifth step involves assigning the values ​​of the local accounts to the accounts in the report and generating the report. A program that causes a computer to execute something.

Images