KR101084440B1 - Automatic entry generation appartus and method thereof - Google Patents

Abstract

The present invention automatically obtains the attributes necessary for accounting journal processing by using the character reader in various evidences, and if the evidence is stored as electronic data, it automatically acquires the necessary attributes directly from the stored data to fit the double entry bookkeeping system. The present invention relates to an automatic journal generating device and a method for automatically processing journal entries. According to the present invention, in the case of evidence stored in evidence or electronically stored proof, a raw transaction database storing characters retrieved from stored data and extracted attributes; An accounting subject database that stores accounting subjects corresponding to the attributes; A character reader for reading a character from proof and storing the character in the original transaction database; An attribute extraction unit for extracting an attribute from the character read by the character reading unit and storing the attribute in the raw transaction database; An attribute inspecting unit inspecting attributes extracted by the attribute extracting unit and identifying and reporting attribute suitability such as whether there is a duplicate, false evidence or insufficient attribute, or no missing attribute; An accounting subject assigning unit for assigning an accounting subject using the attribute combination extracted by the attribute extracting unit by referring to the accounting database; And a journal entry completion unit for completing a journal using attributes extracted from an account subject and an attribute extracting unit granted by the account subject granting unit.

Journal Entry, Evidence, Double Entry Bookkeeping, Automatic Journal Entry, Character Readout, Evidence Attribute, Attribute Combination

Description

Automatic entry generation device and its method {Automatic entry generation appartus and method

According to the present invention, when a character reader is used in various evidences, or when evidence data is stored in an electronic form, an automatic journal occurrence is automatically obtained by acquiring the necessary attributes for accounting journal processing by automatically reading them and conforming to the double entry bookkeeping system. An apparatus and a method thereof are provided.

The accounting activities of an entity cause changes that increase or decrease in assets, liabilities, capital, revenues, and expenses. In order to account for such information, bookkeeping is the bookkeeping that divides transactions into debits and credits in books. Journal entries follow some rules.

In other words, debit should include an increase in assets, a decrease in debt, a decrease in capital, and an incurred cost, and in a credit, a decrease in an asset, an increase in debt, an increase in capital, and an occurrence of revenue.

However, in order for the journal entry to be accurate, only a person with expertise in accounting must do this, and also, the person analyzes the accounting activities of these companies and decides the accounting subject, and the accounting subject is debited. After deciding whether to vote or credit, it was inconvenient to write it in a book by hand or input it into an existing accounting program.

In addition, in order for journal entries to be automatically generated, recognition targets, which are the basis for identifying transactions, must be determined, but studies on them have been insufficient. In other words, no logical conclusion can be drawn from "What should be recognized to complete accounting input automatically?" When a transaction occurs, there is evidence (or sometimes referred to as a "seal") that proves the transaction as the exchange of goods, services, consumption, or movement occurs, but the conventional approach does not use objective evidence as an object of recognition. Because of the object of recognition, the mechanical recognition could not be achieved and the manual journal entry method through the judgment of a person familiar with accounting could not be escaped. In addition, most of those currently patented under the name of automatic journal entry are in fact at the level of manual journal entry. Therefore, intuitively, any proof has been considered to be automatic accounting if processed using an optical recognition device.

In this regard, as an automatic journaling application filed so far, there was a "document automatic journaling method and apparatus thereof" of Korean Patent No. 10-419263.

The registered patent allows manual entry of transaction dates, details, etc., and extracts the appropriate account subjects by comparing the details of the manually entered natural language sentences with keyword morphemes in the embedded database. There is some distance from automation as required.

In addition, there is a cyber book management system having an automatic account setting function of the invention of Korean Patent Publication No. 2009-43803.

The disclosed patent is a method applied in the context of fasting bookkeeping, such as cash income and expenditure, which is a problem in that most accounts of corporations and the public today assume double entry bookkeeping. There is a problem with the manual method.

Normally, manual work must be excluded in order to automatically create a journal entry (1) the transaction that is the subject of the journal entry must be automatically recognized first as an objective and specific object, and (2) the content recognized in (1) is generally recognized. It should be converted into account subject, date, amount, and other details to meet the accounting standard. And (3) The result generated in (2) should be expressed in the journal form. Among them, the recognition object of (1) must be something that can be recognized by the machine, so it must be concrete that does not require human judgment.

If you do not target specific and physically identifiable transaction evidence, the object to be recognized must be abstractly identified. Therefore, it is necessary to go through manual recognition by a person familiar with accounting.

However, the evidence is not intended for accounting, but for transactions, and does not contain all the information necessary for accounting. There is some information necessary for journal processing, but there is a lot of unnecessary information, and there are many cases where additional information needs to be supplemented. Nevertheless, no research has been done on which part of the evidence generated from the transaction is necessary for the journal entry and what information is added to complete the journal entry for the double entry bookkeeping.

In other words, one evidence and one account subject do not necessarily have a one-to-one correspondence. For example, if you consider the proof of “tax invoice (for storing recipients),” it can be dozens of account subjects, including goods, products, consumables, R & D, etc., depending on which department receives which item for what purpose. Because.

Therefore, by empirically analyzing the contents of the evidence, it is necessary to examine what information attributes the evidence consists of, and to distinguish between attributes necessary for journal entry and unnecessary attributes, and to supplement additional information attributes necessary for journal entry. Research is needed, and so far no research has been done on this. The problem does not stop here.

Next, due to the special journalistic characteristic of double entry bookkeeping, transactions must be recognized at the same time as the source of the funds and the debit of the management of the funds.There is no consideration of how it should be done. The risk of becoming an incomplete journal entry of ten thousand journals is another problem that must be overcome.

On the contrary, given the fact that more than one piece of evidence occurs in a transaction, there is a risk of the same journal being double or triple.

Furthermore, in today's environment where fraudulent accounting is prevalent, there is a risk that fraudulent accounting occurs automatically if an automatic journal entry is made without the provision of false evidence or missing evidence.

If the above problems are not solved at the same time, it is difficult to say that a true automatic journal entry has occurred.

In addition, given the trend of social electronic receipts becoming more common, such as corporations mandatory to issue electronic tax invoices from individual companies since 2010 and individual tax operators from 2011, the evidence already stored in electronic form need to go through the character recognition process. However, in order for it to be automatically journalized, you must go through the steps as shown above.

However, the above-described patents are still insufficient in this regard, and therefore, a solution of such a problem is required.

In order to solve the above problems, the present invention is to import evidence stored in electronic form, or to obtain the attributes required for accounting journal processing using a text reader from the evidence or electronic evidence file already stored in electronic form. It is an object of the present invention to provide an automatic journal generating device and a method for automatically processing a journal entry.

The present invention for achieving the above object, the raw transaction database for storing the characters and the extracted attributes read in the evidence; An accounting subject database that stores accounting subjects corresponding to the attributes; A character reader for reading a character from proof and storing the character in the original transaction database; An attribute extracting unit for extracting an attribute from the character read by the character reading unit and storing the attribute in the raw transaction database; an accounting subject for granting an account subject using the attribute combination extracted from the attribute extracting unit by referring to the accounting database; Grant unit; And a journal entry completion unit configured to complete the journal using the attributes extracted from the account and the attribute extraction unit provided by the account course granting unit.

In addition, the present invention is characterized in that it further comprises an attribute inspector for inspecting the attributes extracted by the attribute extractor to determine whether there is a duplicate, false evidence, or insufficient attributes to exclude inappropriate attributes or evidence. .

In addition, the character reading unit of the present invention is characterized in that to extract a character from the evidence file stored in the electronic form, the raw transaction database characterized in that for storing the characters extracted from the evidence file stored in the electronic form in the character reader do.

In addition, the attribute extracted by the attribute extractor of the present invention is characterized in that it comprises a document title attribute, item attribute, department attribute, the target attribute.

In addition, the attribute information extracted by the attribute extractor of the present invention is characterized in that it further comprises a customer attribute, the amount attribute, the payment method attribute.

In addition, the account subject providing unit of the present invention is characterized by assigning an account subject by referring to similar terms when the same proof title does not exist in the account subject database.

In addition, the present invention comprises the steps of (A) the character reading unit to read the character from the evidence stored in the raw transaction database; (B) an attribute extracting unit extracts an attribute from the character read by the character reading unit and stores the attribute in the raw transaction database; (C) an account subject assigning unit granting an account subject using the attributes extracted by the attribute extracting unit by referring to the account subject database; And (D) journal entry completion unit characterized in that it comprises a step of completing the journal using the attributes extracted from the account and attribute extraction unit granted by the account subject grant unit.

In addition, the present invention comprises the steps of (E) the character reading unit to retrieve the stored data in electronic form to store in the raw transaction database; And (F) extracting the attribute from the document data stored in the electronic form by the attribute extracting unit and storing the attribute in the raw transaction database.

In addition, the present invention (G) attribute inspector to inspect the attributes extracted by the attribute extractor to determine whether there is a duplicate, false evidence or insufficient attributes, missing attributes to exclude the inappropriate attribute or evidence It further comprises a step.

In addition, the step (C) of the present invention, (C-1) the account subject assigning unit searching for the same evidence attribute combination in the account subject database to search for the presence of the same evidence attribute combination; And (C-2) the account subject assigning unit searching for the same evidence attribute combination in the account subject database, and if the same evidence attribute combination exists, refer to the corresponding account subject and use the attributes extracted from the attribute extracting unit. Characterized in that comprises the step of giving.

In addition, the present invention (C-3) the account subject grant unit search for the same title in the account subject database, if the same title does not exist to search for a similar title there is an account corresponding thereto, the attribute extraction unit with reference to this The method may further include granting an account subject using the extracted attributes.

1) Today, business investors, creditors, managers, information, unions, and consumers are very interested in the financial statements of a company. This is because it is necessary to make decisions after understanding the soundness of corporate management through financial statements. Therefore, prompt financial statements should be provided to these stakeholders. But the reality is not so. In order for a financial statement to be created, all accounting evidence must be gathered to the accountant, and after that, a manual process of assigning appropriate accounting subjects is required. Even in the case of ERP-promoted companies, major tasks such as sales and production are performed by hand in the corresponding departments, so the parallelization and decentralization of accounting input has been achieved, but the final input must be done manually in the accounting department. In addition, electronic receipts are being widely disseminated, such as the issuance of tax invoices in electronic form, but accounting alone has not exceeded the level of manual work. Therefore, regular financial statements are generally generated only two to three months after the settlement date. However, according to the present invention as described above, since the approved evidence can be made up to the financial statements at the same time through the character reading device, there is an effect that the speed of supplying the financial statements to the stakeholders is very shortened.

2) The costs for hiring and training personnel, supervisors and external accounting audits are the greatest annually. However, according to the present invention as described above, only a minimum number of monitoring personnel is required, thereby reducing labor costs and efforts for accounting input.

3) The financial statements of many companies today are questioned for credibility through the use of prudent accounting. The reason for this is that they are greatly distorted by the subjective ability and intention of the inputter, such as inputting a fictitious transaction without proof or inputting an intentional or error different from the evidence. To eliminate these risks, companies have to pay huge internal control costs, while auditing charges rise, but reliability remains questionable. It's because they don't believe in the scorecards that outside objective third parties have scored on their own, rather than measuring them. According to the present invention as described above, since the journal entry is made automatically mechanically, such intentional and unintentional errors or injustices will be greatly reduced, and if the validity of the procedure is confirmed, there is an effect of obtaining higher reliability than in the case of manual work. .

Hereinafter, an automatic journal generating apparatus and a method thereof according to a preferred embodiment of the present invention will be described in detail with reference to FIG. 1.

1 is a block diagram of an automatic journal generating device according to an embodiment of the present invention.

Referring to FIG. 1, the automatic journal generator according to the preferred embodiment of the present invention includes a character reader 101, an attribute extractor 102, an attribute inspector 103, an account subject granter 104, and a journal entry completion. The unit 105 and the output unit 106 are provided.

The character reading unit 101 receives an evidence document from a user and reads and outputs the character described in the document. The character reading unit 101 preferably uses an optical character reading device (OCR). In addition, the character reading unit 101 stores the supporting document in the raw transaction database 110 in the form of a picture file, a text file, an Excel file, a word file, and the like.

In addition, the attribute extracting unit 102 extracts an attribute required for accounting journal processing from the characters described in the document document read by the character reading unit 101 so as to correspond to the document stored in the raw transaction database 110 so as to correspond to the original transaction database. Save to 110, and output. If the supporting information is stored in the electronic form, the necessary character is extracted from the attribute extracting unit and the same process as above is performed. Here, the attributes necessary for the accounting journal processing include a date, a document title, a transaction item, an amount of money, and a payment method.

Here, evidence refers to a document which proves the occurrence of a transaction and includes important attribute information in the content of the transaction.

For example, tax invoices, bills, orders, invoices, packing lists, etc., when sales take place. This evidence is very diverse and varies from company to company and from transaction to transaction. Therefore, various classifications are possible, but here are classified as follows for convenience.

1. Proof of sales, purchase transaction

Tax Invoice, Invoice, Simple Receipt, Credit Card Receipt, Cash Card Receipt

Sales agreement, lease agreement, bills, etc.

This evidence relates to the transaction concerning the generation of revenue and the generation of down payments in debt in the case of proof of sales. At the same time, the payment method information is provided as evidence of the generation of trade receivables (assets). In the case of proof of purchase, there is evidence of the acquisition of assets and the incurring of costs. At the same time, the payment method information is provided to prove the occurrence of trade payables (liabilities).

2. Proof of labor cost

-Payroll book, Job account book, withholding receipt, payment statement, etc.

This evidence is not only sales management expenses or labor cost incurred cost of sales, but also information on related liabilities (unpaid expenses) according to the payment method information.

3. Proof of Financial Transactions

-Deposit Account, Loan Account, Income Interest Payment Statement, Payment Interest Receipt

Invoice, wire transfer receipt, bank transfer receipt, foreign exchange transaction receipt

-Copy of stock transaction account book, etc.

Proof of financial transactions is evidence of increasing and decreasing transactions, such as deposits, securities, investments, and borrowings, which provide information that has a broad impact on the accounting of all transactions.

4. Evidence of Capital Transactions

-Registered copy, certificate of payment of receipt, receipt, bill of attorney, etc.

Board of Directors or General Meeting

5. Proof of Cash Transactions

-Receipt, shipping receipt, inventory inspection statement, etc.

7. Proof of Other Cost Transactions

-Courier receipt, simple receipt, ticket

8.Proof of Settlement Specification

Year-end stock price data sheet, year-end exchange rate data sheet, etc.

These evidences are not evidence that arises from transactions with external third parties, but evidence that must be collected and entered artificially for the purpose of settlement. In other words, it must be supplemented to meet the requirements of accounting standards, such as the end-of-life evaluation of asset liabilities and the calculation of the periods of income and expenses.

The reason for starting a journal entry in this evidence is to: 1) prevent entry of unprocessed or over-presented transactions, 2) omit real transactions, or prevent entry of under-represented transactions, 3) Journal entries should be based on evidence to prevent errors or deliberate manipulation of personnel.

However, if all supporting data should not be subject to accounting input, the following principles should be observed:

1) All transactions shall be proved by evidence. (However, in the case of period-end closing journals, such as depreciation or provisioning, actual physical transactions do not occur. In this case, evidence must not be generated outside the company.

2) All evidence of transactions should be approved by the superiors according to the internal control process.

3) Even if the documentary evidence is approved, it shall be a document that is recognized as appropriate for accounting input by understanding the internal control flow of the company.

For example, in sales transactions, there are several steps, such as order receipt, credit approval decision, order release, issue and collection. Each process is subject to internal controls such as personnel, supervisors and internal auditors.

Entering all of these evidences will result in the error of multiple entries of a single transaction multiple times, so it is necessary to evaluate the flow of internal controls in advance and enter the most appropriate accounting standards. In such a case, it is generally best to input the transaction that is generated once according to the tax invoice and then enter the collection transaction through the receipt when it is collected. As such, a company's transaction must predetermine appropriate evidence in consideration of its recurring nature and its uniqueness. And input must go through an internal control procedure that must be approved. Otherwise automation of input will only cause confusion.

These evidences are generally recognized as accounting evidences suitable for accounting input in the enterprise. Although these types of evidences are very diverse, their contents can be expressed as follows.

1) The title attribute of the evidence includes information about the subject of the transaction and the type of transaction.

For example, in the case of a tax invoice (for holding a supplier), it means a sale, sale or service provided by the company. In another example, in the case of an electricity bill, it means a transaction in which the company received a power service. Among account subjects, the scope is compressed into power cost (sales management cost) or power cost (manufacturing cost).

2) Date attribute of the transaction: Contains the time information of the transaction. However, there is no function to reduce the scope of accounts and subjects. For example, the date of a transaction statement is the date on which the goods or services are provided or the transaction received. In another example, the date of the payment slip is the day on which the payment transaction takes place.

3) Other party's attribute of transaction: It contains information about the object (item) of the transaction. The counterparty typically has a significant impact on which item and what type of transaction to pay for in a company. Therefore, the counterparty information is very important to reduce the scope of the account subject. For example, it provides transaction item information provided in the case of supplier personal information of a restaurant simple receipt. The scope of account subjects is compressed into benefits and expenses (sales management expenses or manufacturing expenses).

4) Attributes of transactions (including quantities and unit prices) Attributes: Because they contain information about the subject of a transaction, this also has a significant effect on reducing the scope of account subjects. In addition, there is often a case that overlaps with the counterpart information of the transaction. In this case, since the transaction fact is pointed out more specifically than the counterpart information of the transaction, it is often necessary to be understood as more respected information when selecting an account subject. For example, in the case of a transaction item, a quantity, and a unit price of a tax invoice (for storing a recipient), the scope of the account subject is compressed into whether the item is a cost of sales or an administrative expense.

5) Transaction amount, tax amount and sum (attribute amount): It contains information on the economic value of the transaction.

6) Method of Payment or Deposit (Payment Method Attributes): Contains value information about the payment of the transaction. This information plays an important role in determining what the other account is. In other words, all transactions must be accounted for in two aspects: (debit) = "operation" and "credit" = "source of funds." If the information in the "debit or credit account" designates, "payment information" generally determines the reverse account. In other words, in the absence of attribute information on the payment method, it is more likely that a journal entry will not occur in the form of double entry bookkeeping. The payment information is formed by matching and extracting corresponding information from transactions having the same date, account, or item information using transaction date information, customer information, and item information.

7) Purpose attribute: Provides relevant information on the purpose of the transaction. As we have seen earlier, although the scope of the accounting subject to display the transaction is reduced by "evidence title", "transaction item" and "trading partner" information, there is still uncertainty that may be determined by more than one accounting subject.

If the historical information is missing or insufficient, there are too many kinds of account subjects that can correspond to a document, and the uniqueness of 1: 1 correspondence cannot be maintained. This is because the final classification criteria "purpose" information is missing in the determination of the account. Therefore supplementary information on this should be provided here. And, "the purpose of the transaction" can be confirmed by clarifying where the department in which the transaction object is used or what the intention of the manager is.

For example, if the tax invoice (for storing recipients) indicates that the department is "for research and development," the appropriate account subjects are compressed into research, development, or ordinary research and development expenses. In addition, if the "information related to the project" is included in the transaction approval process, it is possible to specify exactly one account subject suitable for the project.

8) Department attribute: It is easy to omit in most evidences, and the subjects to be dealt with may differ depending on how the department, the item of transaction, the amount of money, and the amount of money are the same. For example, if you purchased a TV on your tax bill (for storing your recipients), you should be given a different accounting account depending on whether you use it in the sales department, research department, management department, or production department.

On the other hand, the lack of historical or departmental information must already be given before the accountant can judge by hand, so it is reasonable that internal controls are supplemented at the stage of approving the transaction. Historical or departmental information should be given in the process of settlement by the approver who understands and approves the transaction. However, there is a problem that automation is difficult when the details or department information to be given are subject to the judgment of the accountant, so a list of necessary details or department information should be prepared in advance. In this regard, it is necessary to supplement the historical information and the department information by stamping the management code or use code of the department in advance in the transaction approval process.

Next, the raw transaction database 110 includes 1) company basics (business registration certificate, registered copy, shareholder list, employee list, etc.), and 2) customer information (trade name, representative, address, business registration number, business status, item, handling). Items, major account subjects, etc.), 3) transaction items (items, accounts, major account subjects, etc.), 4) documentary registration (main evidence attributes such as titles, accounts, items, amounts, etc.), and 5) transaction details. 6) major contract documents (rental agreement, contract of sale, borrowing agreement or summary), 7) settlement transaction reference data (term ending stock quotes, period exchange rate table, depreciation data, etc.), 8) similar terms, etc. Equipped.

In this case, the main accounting subjects referred to in 2, 3, and 4 refer to information listing accounting accounts that are most likely to occur. Raw transaction databases containing this information should be updated periodically.

Next, the accounting database 111 is composed of a combination of a plurality of evidence attributes and a plurality of debit and credit courses corresponding to the combination of a plurality of evidence attributes, and consists of the following Table 1. That is, one documentary attribute and one accounting subject are not 1: 1 correspondences, but a combination of sufficient evidence attributes and one accounting subject is a principle that achieves a multi: multiple correspondence.

(Table 1)

Evidence
Attribute 1 Evidence
Attribute 2 Evidence
Attribute 3 Evidence
Attribute 4 debit side
Subject 1 debit side
Subject 2 credit
Subject 1 credit
Subject 2

Here, the documentary attributes refer to attributes that play a decisive role in designating account subjects such as document titles, accounts, trading items, transaction details, and payment methods as described above.

For example, a journal entry for a transaction that meets these four (if necessary, the number of documentary attributes can be increased or decreased further) appears as a debit or credit account, which is shown in the next field. Since it may appear as 1, 2 and so on to add more.

These tables should be prepared in advance. This table also needs to cover all cases where a transaction can occur. Therefore, if an unexpected new type of transaction occurs, it should also be updated.

In addition, the types of transactions to be included in this table must be mutually exclusive and not overlap. Being exclusive means that every case in the representation is unique.

Maximum number of cases where different accounting takes place = (number of evidence types) * (number of registered accounts) * (number of registered transaction items) * (number of payment methods) * (number of history)

For example, if there are 60 types of titles of registered evidence, 80 customers, 90 items traded, 5 methods of payment, and 30 items, the total number of journal entries that can occur is 64800000. . However, it is not necessary to practically prepare so many databases. Because, even though there are about eight attributes in a combination of evidence generated by a transaction, not all of them will affect the finding of a suitable accounting subject with equal weight, and in general, considerably by evidence title attributes. The number of cases no longer needs to be taken into account and the practical experience proves that the number of cases that can actually occur is rapidly duplicated whenever the next attribute is considered. Therefore, in principle, the combination of attributes is considered at the same time, but in practice, it is a realistic procedure to collate the content of the evidence from the database and the transaction evidence.

As such, a table for assigning an account subject is created in the account subject database 111. In such a situation, the account subject grant unit 104 refers to the account subject database 111 and uses the attribute information extracted from the evidence to account. Assign courses.

To this end, the account subject assigning unit 104 first determines whether an attribute combination identical to the attribute combination extracted by the attribute extracting unit 102 exists in the database 111 and finds and assigns an account subject corresponding to the designated debit and credit in that case. do. However, in practice, since the document title among the document attributes often provides preferential information when determining the account subject, it is necessary to check whether the document title database 111 has the same document title for the document title first. If it is determined that the same documentary title exists, the next step is to determine whether there is a journal entry that matches the counterparty, transaction item, payment method, and details.

If there is no matching journal entry, the account subject assigning unit 104 determines whether there is a similar term with reference to the account subject database 111.

The account subject granting unit 104 determines whether there is a journal entry that matches the counterparts, transaction items, payment methods, and details below, when there is a similar term corresponding to the case where there is a similar term for the proof title. 103, if there is a matching journal entry, grants the corresponding debit and credit account.

However, as a result of searching for a similar term, the account subject providing unit 104 notifies the administrator if there is no similar term so that the administrator can update the account subject database 111.

Of course, instead of pre-creating these databased tables and assigning account subjects, in a more general way, a plurality of supporting attributes are combined to form an algorithm to assign corresponding account subjects when a unique transaction occurs. For example, there may be a method of assigning account subjects.

On the other hand, after the account subject grant unit 104 confirms the account subject, the journal entry completion unit 105 grants the transaction date and the amount. At this time, it is determined whether the amount calculation is a separate transaction or not. Calculate the amount and give the calculated transaction amount.

Generally, the transaction amount is temporarily stored in the original transaction database 110 and used for journal entry. This is the case of acquisition, disposal, sale, etc. that occur during the accounting period, in which case the amount of the transaction evidence is used without any additional manipulation.

However, accounting standards often require revaluation of assets, liabilities, depreciation, loan loss provisions, retirement allowances, and foreign currency valuation. In most of these cases, proof of transaction with an external third party does not occur. Such a transaction is called "closing transaction" in accounting, which is an artificially assumed transaction. A separate internal proof must be provided for this settlement. The reason for this is as follows.

(1) Settlement transactions required by accounting standards often require subjective estimates.

For example, in order to establish loan loss provisions, the manager must subjectively estimate the recoverability of the trade receivables, and the manager must subjectively estimate the depreciation method and useful life even if he or she intends to depreciate depreciation. This subjective element is an inherent limitation of accounting, so management must approve the transaction and approve it before the accountant can process the account.

(2) Accounting standards today require that when assessing assets and liabilities, and determining the amount of revenues and expenses, the best estimates of future cash flows should be based on, for example, the quoted prices of invested stocks Tables, depreciation data, etc., are already prepared in the database before the settlement of accounts and the amount is determined.

(3) A closing transaction is an artificially created transaction and does not exist in the database. Therefore, it is necessary to artificially create the closing transaction in the raw transaction database.

As such, in most settlement transactions, the determination of account amounts becomes more important than the determination of account subjects. The amount of the transaction is determined according to the algorithm specified in the accounting standard.

For example, in the case of a terminal stock valuation transaction,

(Debit) Short-term trading securities (Credit) Short-term trading securities

At this time, transaction amount = period stock value-book value of stock

             -(Ending stock value per share * Number of shares held)

Becomes

Next, the journal entry completion unit 105 receives the accounting information from the accounting section granting 104, checks whether there is no error in the information, and if there is an error, provides an error message to the user and an item with an error. Receive data from the user. If there is no error as a result of the determination as above, it determines the accounting document information according to the assigned account subject and serves to store in the accounting journal database 112.

The output unit 106 does not extract the account subjects from the account subject granting unit 104, or if a plurality of extracted ones is not confirmed, displays a question to confirm this and provides it to the user, journal entry completion unit In 105, if there is an error in the account subject, an error message is displayed and displayed to the user.

In addition, the output unit 106 outputs the completed journal entry result to the journal completion unit 105 in the form of a slip.

On the other hand, the attribute inspecting unit 103 is located at the rear end of the attribute extracting unit 102, by inspecting the attributes extracted by the attribute extracting unit 102, whether there is a legitimate transaction, whether there is a double input, whether there is a missing transaction Etc. are inspected through other manageable means such as serial number or digital check to provide the account subject granting unit 104 with the attributes extracted by the attribute extracting unit 102, excluding inappropriate attributes or evidence. This allows for more accurate journal entry processing. At this time, the attribute inspecting unit 103 outputs an error message if an inappropriate evidence or an attribute is found so that it can be re-injected by correcting the evidence information stored in the original or electronic form by receiving a new approval from the superior according to the internal control system. .

2 is a flowchart of an automatic journal generation method according to a preferred embodiment of the present invention.

Referring to FIG. 2, in the automatic journal generation method according to the preferred embodiment of the present invention, the character reading unit 101 first receives a document from the user and outputs the document described in the document. The character reading unit 101 preferably uses an optical character reading device (OCR). In addition, the character reading unit 101 stores documentary evidence as a picture file. In addition, the character reading unit 101 reads a character from the stored evidence data and stores it in the raw transaction database 110 when the proof data is stored in the form of electronic data.

Next, the attribute extracting unit 102 extracts and outputs an attribute necessary for the accounting journal process from the characters described in the document which is read by the character reading unit 101 (S112). In this case, the attributes extracted by the attribute extracting unit 103 include a document title (evidence title attribute), a date, a counterparty, a transaction item (item attribute), an amount of money, a tax amount, a payment method, a purpose, a department, and the like.

Next, the attribute inspecting unit 103 is located at the rear end of the attribute extracting unit 102 and inspects the attributes extracted by the attribute extracting unit 102 to determine whether there is a legitimate transaction, whether there is a double input, or whether there is a missing transaction. The presence or the like is inspected through other manageable means such as serial number or digital check, and the attribute extracted from the attribute extracting unit 102 is provided to the account subject granting unit 104 except for inappropriate attributes or proofs (S113). ). At this time, the attribute inspecting unit 103 may output an error message if an inappropriate evidence or attribute is found, correct the evidence information stored in the original proof or the electronic form according to the internal control system, and re-inject it. (S114).

Subsequently, the account subject assigning unit 104 determines whether there is the same supporting attribute combination in the account subject database 111 with respect to the document title among the attributes extracted by the attribute extracting unit 102. If it is determined (S115), it is determined whether there is a journal entry that matches the transaction counterpart, the transaction item, the payment method, and the details (S116).

And, if there is a matching journal, the account subject grant unit 104 grants the corresponding debit and credit account subjects (S117).

If there is no matching journal entry, the account subject grant unit 104 refers to the raw transaction database 110 to determine whether there is a similar term (S118).

The account subject assigning unit 103 determines whether there is a journal entry that has a similar term corresponding to a case where there is a similar term for a proof title and a transaction counterpart, a transaction item, a payment method, and a description match (S116).

And, if there is a matching journal, the account subject grant unit 104 grants the corresponding debit and credit account subjects (S117).

However, as a result of searching for similar terms, the account subject providing unit 104 notifies the administrator when there are no similar terms, so that the administrator can update the account subject database 111 (S119).

Meanwhile, after the account subject grant unit 104 confirms the account subject, the journal entry completion unit 105 grants the transaction date and amount (S120) to complete the journal entry (S121). If it is determined whether or not (S122) the amount calculation is necessary to calculate the transaction amount is given to the calculated transaction amount (S123).

FIG. 3 is a detailed flowchart illustrating the process of determining the accounting subject of FIG. 2. First, debit and credit are classified using the document title attribute (S210).

Then, assets, liabilities, capitals, revenues, expenses, etc. are classified using the document title attribute (S212).

Subsequently, although not shown, additionally, asset liabilities are classified as current and non-current assets, based on a one-year period of asset liabilities.

Next, in the case of classifying the asset, the realization method of the asset is classified using the item attribute and the target attribute, in addition to the title attribute, and classified into cash, receivables, inventory assets, and tangible assets (S220).

On the other hand, when classified as a debt, in addition to the title attribute, classifying the occurrence type of the debt by using the item attribute and the target attribute and classifies it as a payable note, retained earnings (S230).

In addition, when the capital is classified as capital, the occurrence type of capital is classified using the item attribute and the target attribute, in addition to the title attribute, and classified as capital, retained earnings, etc. (S240).

Next, when classified as revenue, using the item attribute, department attribute and purpose attribute other than the title attribute to determine the main sales relevance of the revenue and classify it into revenue and non-zero revenue (S250).

Subsequently, by classifying the generation type of revenue using the item attribute, the department attribute, and the target attribute other than the title attribute, it is determined whether the product sales or the product sales (S252).

On the other hand, if it is classified as a cost, by using the item attribute, department attribute, and purpose attribute other than the title attribute to determine the main business relevance of the cost, it is divided into cost of sales, SG & A expenses, and non-English expenses (S260).

Subsequently, by classifying the occurrence department of the cost by using the item attribute, department attribute and purpose attribute other than the title attribute to determine whether the manufacturing cost or construction cost (S262).

In addition to the title attribute, the occurrence of expenses is classified using the item attribute, department attribute, and purpose attribute to determine whether it is a payment fee or a research and development cost (S264).

Next, in addition to the title attribute, the item attribute, the department attribute, and the target attribute are classified to determine whether the expenses are incurred or whether they are entertainment or welfare benefits (S266).

1 is a block diagram of an automatic journal generating device according to an embodiment of the present invention.

2 is a flow chart of the automatic journal generating method according to an embodiment of the present invention.

FIG. 3 is a detailed flowchart of the process of determining the accounting subject of FIG. 2.

<Explanation of symbols for the main parts of the drawings>

101: character reader 102: attribute extractor

103: attribute checking unit 104: accounting subject granting unit 105: journal entry completion unit 106: output unit

109: Evidence file stored in the form of electronic data

110: raw transaction database

111: Accounting Database 112: Accounting Journal Database

Claims (15)

A raw transaction database storing characters read from the evidence and extracted attributes; An accounting subject database that stores accounting subjects corresponding to the attributes; A character reader for reading a character from proof and storing the character in the original transaction database; An attribute extraction unit for extracting an attribute from the character read by the character reading unit and storing the attribute in the raw transaction database; An accounting subject assigning unit for assigning an accounting subject using the attribute combination extracted by the attribute extracting unit by referring to the accounting database; And Automatic journal generation device comprising a journal entry completion unit for completing the journal using the attributes extracted from the account and attribute extraction unit granted by the account subject grant unit. The method according to claim 1, And an attribute checking unit for checking whether the attribute is extracted by the attribute extracting unit to determine whether there is a duplication, false proof, or insufficient attribute, thereby excluding an inappropriate attribute or proof. The method according to claim 1, The character reading unit may extract a character from the evidence information stored in the electronic form, The raw transaction database is an automatic journal generating device, characterized in that for storing the characters extracted from the evidence information stored in the electronic form in the character reading unit. The method according to claim 1, The attribute extracting unit extracts an attribute including a document title attribute, an item attribute, a department attribute, and an objective attribute. The method according to claim 4, The attribute information extracted by the attribute extractor further includes an account attribute, an amount attribute, and a payment method attribute. The method according to claim 1, And the account subject assigning unit assigns an account subject by referring to a similar term when the same proof title is not present in the account subject database. The method according to claim 1, The account subject grant unit categorizes the debit and credit using a document title attribute combination. The method according to claim 1, The accounting subject assigning unit classifies the asset realization method using a title attribute, an item attribute, and a target attribute, classifies the generation type of debt, and classifies the rights obligations of capital. The method according to claim 1, The account subject providing unit uses the title attribute, item attribute, purpose attribute, and department attribute to determine the main sales relevance of revenue, revenue generation type, sales relevance of expenses, cost incurring items, cost incurring department, and purpose of incurring expenses. Automatic journal generating device characterized in that the classification. (A) the character reading unit reads the character from the evidence and stores it in the raw transaction database; (B) an attribute extracting unit extracts an attribute from the character read by the character reading unit and stores the attribute in the raw transaction database; (C) the account subject assigning unit grants the account subject using the attribute combination extracted by the attribute extractor by referring to the account subject database; And (D) journal entry completion unit comprising the step of completing the journal using the attributes extracted from the account and attribute extraction unit granted by the account subject granting unit. The method according to claim 10, (E) a character reading unit to retrieve the evidence stored in the electronic form and to store in the raw transaction database; And (F) an automatic journal creation method further comprising the step of extracting the attribute from the document stored in the electronic form in the text reader in the raw transaction database. The method of claim 10, (G) inspecting the attributes extracted by the attribute extracting unit to determine whether there is a duplicate, false evidence or insufficient attribute, and whether there are no missing attributes to exclude the inappropriate attribute or evidence; How automatic journals occur. The method according to claim 10, Step (C) is (C-1) searching for an identical subject by searching for an identical subject in the account subject database by an account subject assigning unit; And (C-2) the account subject granting unit searching for the same subject in the account subject database and if the same subject exists, granting the account subject using the attribute combination extracted from the attribute extractor by referring to the corresponding account subject. Automatic journal generation method comprising a. 14. The method of claim 13, (C-3) If the same subject does not exist by searching for the same document attribute in the account subject granting unit, there is an account corresponding to the similar document attribute, which is extracted by the attribute extracting unit. Automatic journal creation method comprising the step of assigning the accounting subject using the attribute combination. 14. The method of claim 13, Step (C-2), Classifying the debit and credit using the document title attribute; Classifying assets, liabilities, capital, revenues, expenses, etc. using documentary title attributes; Classifying an asset realization method using an item attribute and a target attribute in addition to the title attribute when classified as an asset; Classifying the occurrence type of the debt by using the item attribute and the target attribute in addition to the title attribute when classified as a debt; Classifying the occurrence type of capital using the item attribute and the target attribute in addition to the title attribute when classified as capital; When classified as revenue, determining the main sales relevance of the revenue using item attributes, department attributes, and destination attributes other than title attributes; Classifying a generation type of revenue using item attributes, department attributes, and purpose attributes in addition to title attributes when classified as revenue; If it is classified as a cost, determining the main sales relevance of the cost by using an item attribute, a department attribute, and a purpose attribute other than the title attribute; Classifying the occurrence department of the cost using the item attribute, the department attribute, and the target attribute in addition to the title attribute when classified as the expense; Classifying occurrence items of expenses using item attributes, department attributes, and purpose attributes in addition to title attributes when classified as expenses; And A method of generating automatic journals, comprising classifying the purpose of incurring expenses using item attributes, department attributes, and purpose attributes in addition to title attributes when classified as expenses.

Images