KR20030015708A - Method of automatically journalizing for slip and apparatus thereof - Google Patents

Abstract

PURPOSE: A method for automatically journalizing a slip and a device thereof are provided to automatically journalize accounting activity information inputted as a natural language form. CONSTITUTION: Transaction details information is received as a natural language sentence(101). The transaction details sentence is analyzed according to a part of speech including a subject, an object, and a verb(102). A main activity element is extracted based on the verb of the analyzed sentence(103). A basic attribute of the main activity element is extracted in accordance with a use or contents of the extracted main activity element(105). A debtor account item and a credit account item are extracted in accordance with the basic attribute of the main activity element(107,109). Information of necessary items according to the account items is created based on the transaction details information(111). Accounting slip information is generated based on the extracted account items and information of necessary items(112-114).

Description

Method of automatically journaling and apparatus thereof

The present invention relates to a method and apparatus for journaling accounting information, and more particularly, to a method and apparatus for receiving accounting activity information in the form of natural language and automatically journaling it.

A company's accounting activities cause changes in assets and liabilities to increase or decrease, and to account for this information, the bookkeeping is divided into debits and credits in the bookkeeping, which is called a journal entry. do. In other words, debit should include an increase in assets, a decrease in debt, a decrease in capital, and incurred expenses.

However, in order for the journal entry to be accurate, there is a problem that only a person with expertise in accounting needs to do this. Also, a person analyzes the accounting activities of these companies in order to set accounting subjects and debit the accounting subjects. After deciding whether or not to be credited, it was inconvenient to write it in a handbook or input it into an existing accounting program. Because of this inconvenience, those who do not have accounting knowledge have no choice but to hire and handle accounting managers or entrust them to professional institutions (accounting offices / tax offices).

The present invention to solve the above problems is to provide a method and apparatus for automatically journaling accounting activity information input in the form of natural language.

1 shows an overall flowchart of a preferred embodiment of an automatic journaling process according to the present invention.

FIG. 2 shows a flowchart of a preferred embodiment of the detailed process of extracting the main activity factor of FIG. 1.

3 shows a flowchart of a preferred embodiment of the detailed process of extracting the debit and credit account subjects of FIG. 1 (107 to 110).

4 shows a flowchart of a preferred embodiment of the detailed process of the error checking step 113 of FIG.

Figure 5 shows a block diagram of a preferred embodiment of an automatic journal device according to the present invention.

The automatic journal method according to the present invention for achieving the technical problem to be achieved by the present invention comprises the steps of (a) receiving the accounting activity information in a natural language sentence, (b) giving a sentence of the accounting activity information, including the object, verb Analyzing main parts of speech, (c) extracting main activity factors based on verbs of the analyzed sentence, and (d) basic attributes of the main activity factors according to the purpose or contents of the extracted main activity factors (E) extracting an account subject according to basic attributes of the main activity factor, (f) generating information of necessary items for each account subject based on the accounting activity information, and (g) the And generating accounting document information on the basis of the extracted account subjects and necessary item information. In addition, the step (c) is the step of (c1) registering the main activity factor to distinguish the contents of the accounting activity and (c2) determine the main activity factor based on the registered main activity factor and the verb of the analyzed phrase And (c1), registering the main activity factor including priority information of the main activity factor, and (c3) the main activity factor extracted in step (c2). The method may further include extracting main activity factors of the input transaction details based on the registered priority information. In addition, step (c1) above includes a questionnaire for selecting the main activity factor in relation to the registered main activity factor, and (c4) if there is no main activity factor determined in step (c2), Providing a question item to the user, and receiving the answer to the question item may be characterized in that it further comprises the step of determining the main activity factor based on the answer, the automatic journal method above (h) the Determine whether the sum of the debit amount and the credit amount of the generated accounting document is the same, whether the expense account has been debited only and the profit account has been made only on the credit, checking whether the accounting document has an error. The method may further include determining document information.

The automatic journal device according to the present invention for achieving the technical problem to be achieved by the present invention is an input unit for receiving the necessary information during the journal entry processing of the natural language sentences and automatic journals related to the accounting activity, analyzing the natural language sentences for each component A syntax analysis unit extracts the main activity factor and basic attribute information of the main activity factor according to the components of the analyzed natural language sentence, and then extracts the debit and credit account subjects according to the contents of the main activity factor and the basic attribute information. Automatic journal unit for confirming the accounting documents by journal entry processing transaction; And an output unit for providing the user with information generated during the journal entry processing process of the automatic journal unit and providing the determined accounting information to the user. The automatic journal unit may include the output of the analyzed natural language sentence. A main activity factor extracting unit for extracting a main activity factor based on a verb and extracting the main attribute of the main activity factor according to the purpose or content of the extracted main activity factor, and a debit and It may be characterized in that it comprises an account subject extracting unit for extracting the account subject of credit, the accounting information generated by processing the transaction history journal entry information and the sum of the debit amount and the sum of the credit amount of the generated accounting information is the same Whether the expense account has been debited only and the revenue account has been credited only It may be characterized in that it further includes an error check and a document confirmation to check whether there is an error in the accounting document, and confirm the accounting document only when there is no error in the inspection.

DETAILED DESCRIPTION Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. FIG. 1 shows an overall flowchart of a preferred embodiment of an automatic journaling process according to the present invention. First, the accounting activity history is input 101 in the form of natural language, and the syntax of the received natural language sentence is analyzed 102. At this time, referring to a database (syntax structure information DB) that stores information on the structure of the syntax in advance, the words of the natural language sentence are classified and analyzed for each part-of-speech or role. As a result of syntactic analysis, the natural sentence is largely derived from verbs and other parts of speech (nouns, adjectives, adverbs, etc.) that are subordinate to the verb. In this way, verbs and other parts of speech that are dependent on them are represented by their own component markers (English symbols), representing their specific constituents (tubular nouns, adjective nouns, linked verbs, adverb verbs, etc.). It is used to derive which accounting activities belong to, accounting subjects, amounts by accounts, and necessary items by account (quantity, unit price, customer, transaction date, etc.). The unique ingredient indicator used here is as follows.

(1) HEAD: Means the ending word (verb), and all words are analyzed around it. It is also a major determinant of accounting activities (purchasing, selling, etc.), like other companies.

(2) SUBJECT: It means the subject (noun), and the subject of the sentence is the subject of accounting activity (person), which can represent the account subject or necessary items (quantity, unit price, customer, transaction date, etc.).

(3) OBJECT: It means an object (noun), and it can represent an account subject or a necessary item as an object of accounting activity (required item-amount of goods or payment when purchasing or selling).

(4) SCOMP: Refers to a bore (noun) and can represent accounting subjects or necessary items related to accounting activities.

(5) NNMODS: It means a tubular noun phrase, and it is a noun that modifies a noun and can represent an account subject or a required item.

(6) NVMODS: Refers to a sub-shot noun phrase, and is a noun that modifies a verb and may represent an account subject or a required item.

(7) NCMPD: Refers to a compound noun and can represent an account subject or a required item.

(8) ADNCL: A verb that modifies a noun phrase and is a key factor in determining accounting activities (purchasing, selling, etc.) like other verbs.

(9) ADVP: Adverb phrase (noun), which is a noun that modifies a verb and can represent an account subject or a required item.

(10) NMODS: Other noun means a noun, a noun that modifies a noun and may represent an account subject or a required item.

(11) VMODS: Other verbs Modified phrases (verbs), which modifies verbs, are key factors that determine accounting activities (purchasing, selling, etc.) like other verbs.

(12) NCONJ: This means a noun connected word and is not used as a connected word.

(13) VCONJ: Equivalent consolidation clause (verb), which is a continuation verb, which is a key factor in determining accounting activities (purchasing, selling, etc.) like other verbs.

(14) VSUBJ: Dependent Connection Dependent clause (verb), which is the subject clause, is the main factor that determines accounting activities (purchase, sale, etc.) like other verbs.

For example, a natural language sentence, "July 30, 2001, purchased two computers for sale from Mt. Kumgang for 2 million won and paid in draft" is analyzed as shown in Table 1.

Analysis Explanation of English symbols HEAD-> paid (under payment) / verb / Main verb + OBJECT-> bill (note) / noun / direct object + VMODS-> from / investigation / Verbal phrases + NVMODS-> Geumgangsansa-ro (Geumgangsansa) / Nouns / Adverb phrases (noun phrases act as adverbs) + NCMPD-> day (2001-07-30) / noun / noun + VCONJ-> (purchase purchase) / verb / Parallel connection + OBJECT-> 2 (2) / noun / direct object + NCMPD-> computer / computer noun noun + ADNCL-> To sell (below) / verb Tube + NVMODS-> Won (won) / noun / Adverb phrase + NMODS-> 2000000 (2000000) / noun / Noun

Next, the main activity factor is extracted based on the verb from the parsed phrase as described above (103). Here, the main activity factor means an activity type that classifies accounting activity by activity content, and is classified into purchasing activity, sales activity, collection activity, payment activity, expense expenditure activity, fund activity, and settlement activity. Main activity factors are pre-stored in a database (main activity factor registration DB) that is registered in advance, and the main activity factor is extracted from the analyzed syntax as described above with reference to this step of extracting the main activity factor. An example of extracting the main activity factors for the syntax analyzed in Table 1 is shown in Table 2.

verb Main activity factor Priority Paying Payment activity 2 Purchase Purchasing Activity One

In this case, as shown in Table 2, when two or more main activity factors exist, one main activity factor is determined according to priority information stored in the main activity factor registration DB.

Thereafter, it is determined whether the main activity factor is determined (104). If the main activity factor is not determined from the information input in the above process, the user is provided with a question to confirm the main activity factor and inputs an answer ( 1041), and according to the above answers, the main activity factor is confirmed (1042). The questions provided to the user to determine the main activity factor may be registered in advance in the database (main activity factor registration DB) in advance and may be provided to the user with reference to this if necessary. Table 3 shows an example of questions to ask the user to confirm these main activity factors.

number Question Content One Purchased goods or other goods from the customer 2 Selling goods, products, or other goods to a customer 3 In case of receiving credit for the price of goods sold to customer 4 In case of paying credit for goods purchased from a customer ... ...

This question is provided to the user and the answer, that is, how many questions the content of the entered accounting activity corresponds to, and the main activity factor is determined accordingly.

If the main activity factor is determined in step 104 above, the basic attribute of the main activity factor is extracted (105). The basic attribute of the main activity factor refers to the use in purchase activities, and whether the purchase is made for sale or use and retention, and whether the sale is a sale of a product or a product or a disposal of fixed or other assets. In the payment activity, the payment activity is used to distinguish whether the payment is made for unpaid payments or for accounts receivable. However, there is no basic attribute to classify expenses and financial activities and settlement activities, and only one representative attribute. Therefore, the basic attribute of the main activity factor is extracted through the phrase indicating the purpose when the main activity factor is the purchasing activity among the contents of the phrases analyzed above, and through the phrase indicating the kind of goods sold in the case of the sales activity. In the case of payment activity, it is extracted through the phrase indicating the target amount to be paid. Table 4 shows an example of these main activity factors and their basic attributes.

activity Usage Basic property Purchasing Activity Sell sale Purchasing Activity Using use

In the present invention, the basic attribute information for each main activity factor is registered in the database in advance, and this can be referred to when the basic attribute of the main activity factor is extracted.

After that, it is determined whether the basic attribute of the main activity factor is extracted and determined in step 105 above. If the basic attribute of the main activity factor is not determined in step 105 above, the basic attribute of the main activity factor is determined. It provides a question to be asked and the answer is received (1061). The basic attribute of the main activity factor is determined based on the received answer (1062). In this case, the questions provided to the user in step 1061 may be registered together when registering the basic attribute in the database (main activity factor registration DB), and may be provided to the user with reference to this. Table 5 shows an example of these questions.

number Question Content One The purchased product is to sell. 2 The purchased product is for use or storage. 3 The item sold is a product. 4 The thing sold is an asset. ... ...

After deciding the basic attributes of the main activity factor through this process, debit and credit account subjects are extracted (107, 109). Here, the account is a special format for calculating and recording the increase and decrease of assets, liabilities, profits, etc. of the same kind or the same name in the ledger of the bookkeeping, and the accounting subject refers to the subject of different accounts, that is, the name of each account.

The debit and credit account classes are extracted based on the main activity factor and the basic attribute information of the main activity factor. Table 6 shows an example of extracting account subjects based on the main activity factor and basic attribute information of the main activity factor.

Main activity factor Basic property debit side credit Purchasing Activity sale product Cash, Notes Received, Notes Payable, Accounts Payable, Payables ... Purchasing Activity use Land, Buildings, Vehicles, Furnishings Cash, Notes Received, Notes Payable, Purchases of Accounts, Payables ...

Debit and credit account information corresponding to the main activity factor and the main attribute information of the main activity factor is stored in the database (account registration DB) in advance, and the debit and credit account courses are extracted by referring to this process. can do. At this time, if there is only one debit or credit account for the debit or credit for the main activity factor and the main attribute of the main activity factor, the debit or credit account subject shall be determined. Accounts for the above debit or credit will be extracted. Table 7 shows an example of the process of determining the debit account subject.

Main activity factor Basic property noun Account Subject Purchasing Activity use Computer 2 Large Fish Commercial Firm No equipment

Table 8 also shows an example of the process of confirming credit account subjects.

Main activity factor Basic property noun Account Subject Purchasing Activity use Computer 2 Large Fish Commercial Firm No No Received Notes, No Payable Notes

After that, it is determined whether the debit and credit account subjects are confirmed (108, 110), and the credit account is searched by searching the database (account subject registration DB) where the account subject is registered based on a noun as shown in Table 8 above. When the subjects are extracted, the notes to be received and the notes to be extracted are extracted. When a plurality of account subjects are extracted and not confirmed as one of the inputted information, the user is asked a question for confirming the debit or credit account subject (1081, 1101). The debit or account subject is confirmed by receiving the corresponding answer (1082, 1102). In this case, the questions provided to the user are registered in the database in advance, and are used with reference to this process. Table 9 shows an example of a questionnaire for extracting debit account accounts.

number Question Content Account Subject One Purchased for sale. product 2 Acquired land. Land of fixed assets 3 Acquired a building. building 4 Acquired vehicle carriers such as cars and buses. Vehicle transport sphere 5 Acquired equipment such as a computer and a desk. Fixtures 6 I bought a book like a book. Book printing costs 7 I bought office supplies. Office Supplies

In addition, Table 10 shows an example of a question for confirming credit account subjects when multiple credit account subjects are extracted as notes to be received and notes to be paid, and Table 11 shows when credit account subjects have not been extracted in step 109. An example of questions to be provided to the user for extracting credit account subjects is shown.

number Question Content Account Subject One The bill you paid was issued by a third party. Note to receive 2 The bill paid is the bill issued by the company. Notes

Main activity factor Basic property Questions Account Subject Purchasing Activity For sale or use 1. You paid in cash. cash sale 2. paid by trauma. Accounts receivable use Payables For sale or use 3. Paid a third party bill. Note to receive For sale or use 4. Paid its bill. Notes

After the debit and credit account subjects are confirmed, the necessary items for each account are matched (111). In this step, items such as amount, customer, date, quantity, unit price, etc. derived from the parsing result of natural language are matched with items defined as necessary for each account subject. It is input directly. Up to now, the necessary items for each account of the example ("Accounting activity history of purchasing two computers for sale for two million won from Kumgang Sangsan on July 30, 2001 for a bill") will be described with reference to FIG. 6 shows an example of matching.

Thereafter, after receiving input 112 of the necessary items that are not derived from the natural language, it is checked whether there is no error in the account subject, debit and credit amount generated in the above process (113). Here, the error checking method checks whether the sum of the debit amount and the sum of the credit amount is the same, and describes the eight principles of journal entry (debit includes increase in assets, decrease in debt, decrease in capital, and incurring expenses, The credit includes a statement of whether the expense account has been debited only and the profit account has been generated only by the credit, due to a decrease in assets, an increase in debt, an increase in capital, and an income. Check for errors.

If the error is checked and there is no error, the document is confirmed (114) and stored in the database (accounting document DB). Later, the document information thus determined becomes basic data reflected in various books and accounts.

FIG. 2 shows a flowchart of a preferred embodiment of the detailed process of extracting the main activity factor of FIG. 1. First, the main activity factor is extracted 201. This process has already been described in detail with reference to step 103 of FIG. Thereafter, whether the main activity factor extracted in step 201 exists or not (202). If there is no extracted main activity factor, the user is provided with a question for determining the main activity factor (206). The answer to the question is input (207) and the main activity factor is determined based on the answer. In this process, the questions provided to the user are provided to the user by referring to the questions stored in the database (main activity factor registration DB) in advance as in FIG. 1. However, if the extracted main activity factor is present, it is determined whether there are a plurality (203). If there are a plurality of main activity factors, the high priority activity is selected as the main activity (204). Such priority information can be used with reference in this process by registering together when registering the main activity factor in the main activity factor registration DB described above. If there are not a plurality of activities extracted in step 203 above, the main activity factor is determined by selecting the extracted activity as the main activity.

3 shows a flowchart of a preferred embodiment of the detailed process of extracting the debit and credit account subjects of FIG. 1 (107 to 110). Here, the process of extracting the debit and credit accounts is shown. First, the accounting subject is extracted from the accounting activity history information of the natural language input in step 101 of FIG. 1 (301). If the number of extracted account subjects is less than 1 by determining whether the number of extracted account subjects is greater than 1 (302), it is determined whether the number of extracted account subjects is equal to 0 (303). The account subjects will be confirmed as debit or credit account subjects. When the number of extracted account subjects is greater than 1 in step 302, the question items for the extracted plurality of account subjects are displayed (305), and the user's response to the question items is input (306). In this case, the questions provided to the user may be registered in advance, and may be referred to and used in this process as described with reference to FIG. 1. In addition, when the number of account subjects extracted in step 303 is 0, the questions about the candidate account subjects are displayed (307), and the user's response to the questions is input (308) to determine the account subjects based on this. In this case, the questions provided to the user are registered in advance, and can be used for reference in this process, as described in FIG. 1. When an account subject is extracted through this process, the debit or credit is based on the eight principles of journal entry. The credit account account is confirmed (304).

4 shows a flowchart of a preferred embodiment of the detailed process of the error checking step 113 of FIG. First, it is checked whether the sum of the debit amounts is equal to the sum of the credit amounts (401). If the sum amounts are the same, it is determined whether all necessary items are registered (402). It is determined whether there is an error in the journal entry of the account subject (403). Here, the method of determining the error of the account subject is performed based on the eight principles of the journal entry described above as in the case of FIG. On the other hand, if the sum of the debit amount and the sum of the credit amount is not the same in step 401 above, provides an error message to the user and receives the corrected amount (406). After that, the process moves to step 401 again. In addition, if it is determined in step 402 that all necessary items are not registered, an error message is provided to the user (407), and the required items are input (408). Then go back to step 402. If there is an error in the journal entry in step 403, an error message is displayed (409), and the account is input (410) to correct the account. After that, the process moves back to step 403.

Figure 5 shows a block diagram of a preferred embodiment of an automatic journal device according to the present invention. The input unit 501 receives the accounting activity history from the user in natural language form and transmits it to the parser 502, and fails to extract or determine the main activity factor, basic attribute information of the main activity factor, an account subject, and a required item. If it fails to receive the input from the user, the main activity factor and the basic attribute extraction unit 503, the account subjects and necessary items extraction unit 504 or error checking and slip reconfirmation 505, respectively, When adding or modifying data in the main activity factor registration DB, accounting object registration DB, and accounting document DB managed by the device, this information is also input and reflected in the corresponding database.

The syntax analyzing unit 502 receives the accounting activity information in the form of natural language from the input unit 501, and provides basic information necessary for automatically journaling the accounting document by analyzing the natural language sentence for each component. Analyze by constituent above gives a natural language sentence, and analyzes the vocabulary with object, bore, noun phrase, adverb phrase, noun, adverb phrase, noun modifier phrase, verb modifier phrase, noun connection, equal connection subordinate clause, dependent connection subordinate clause, etc. Say what you offer. The analyzed information is transmitted to the main activity factor and the basic attribute extraction unit 503.

The main activity factor and the basic attribute extraction unit 503 receive the phrase analyzed from the parser 502 and refer to the main activity factor registration DB in which the main activity factor is registered in advance based on the verb. If the main activity factor is not extracted or not determined, an answer to the question for determining the main activity factor provided to the user is received through the input unit, and the main activity factor is determined. It plays a role. In addition, the main attribute of the main activity factor is extracted according to the purpose of the extracted main activity factor or the content of the activity. If the basic attribute cannot be extracted, the output element is registered in advance in the main activity factor registration DB to determine the basic attribute. Provides the stored question to the user, and receives the answer through the input unit to thereby determine the basic attribute.

The account subject and necessary item extractor 504 receives the determined main activity factor and basic attribute information of the main activity factor from the main activity factor and basic attribute extractor 503, and extracts the debit and credit account subjects. do. In this case, the accounting subject is extracted by referring to the accounting subject information registered in the accounting subject registration DB based on the main activity factor and the attributes of the main activity factor or the verb or noun of the natural language. In this case, if there are a plurality of extracted account subjects or fails to extract the account subjects, the user is provided with a question for extracting the account subjects through the output unit 506 and the answers are inputted through the input unit 501. It receives the answer to the role and extracts the account subject. In addition, it extracts the necessary item information corresponding to the item of the accounting slip from the accounting activity information in the form of natural language analyzed by the parser 502.

The error checking and slip checking unit 505 receives the accounting subject information and the necessary item information from the accounting subject and the necessary item extracting unit 504, checks whether there is no error in the information, and if there is an error, Provide an error message and receive the data of the item in error from the user. If there is no error as a result of the determination as above, the account item and necessary item information is determined as accounting document information and serves to store in the accounting document DB.

The output unit 506 may not extract each main activity factor or basic attribute information or account subject from the main activity factor and basic attribute extracting unit 503, the account subject and necessary item extracting unit 504, or extract a plurality If one is not confirmed, a question for confirming this is displayed and provided to the user, and if there is an error in the account item or necessary item information in the error checking and the reconciliation of the item 505, an error message is displayed. It serves to display to the user.

6 shows a preferred embodiment of matching necessary items for each account according to the present invention. The document number is composed of 601 and 602 columns, 601 is generated using date information, and 602 can be automatically generated by attaching a serial number. The brief column 603 is a field for receiving accounting activity information in the form of natural language. The item 604 of account number 01 of the account subject is a debit account subject and the amount thereof is matched with 2,000,000 in the debit amount column (606). The paperwork notes 605 of the account subject 02 are credit account subjects, and the amount thereof is matched with 2,000,000 in the credit amount column on the right (607). On the right side of the credit amount column, there is a required item column, which includes "computer" 608 as an item, "2" (609) as a quantity, "20010730" (610) as a receipt date, and "self23400178" (611) as a draft number. It can be seen that "20011020" 612 as the maturity date, "2,000,000" 613 as the amount, and "Kumgangsan superior" 614 as the payment account are matched.

Meanwhile, the above-described embodiments of the present invention can be written as a program that can be executed on a computer. And, it can be implemented in a general-purpose digital computer for operating the program using a medium used in the computer. The media may be stored such as magnetic storage media (e.g., ROM, floppy disk, hard disk, etc.), optical reading media (e.g., CD-ROM, DVD, etc.) and carrier waves (e.g., transmission over the Internet). Media.

So far I looked at the center of the preferred embodiment for the present invention. Those skilled in the art will appreciate that the present invention can be implemented in a modified form without departing from the essential features of the present invention. Therefore, the embodiments disclosed above should be considered in descriptive sense only and not for purposes of limitation. The scope of the present invention is shown in the claims rather than the foregoing description, and all differences within the scope will be construed as being included in the present invention.

According to the automatic journal device or automatic journal method according to the present invention, even those who do not have expertise in accounting can easily prepare an accounting slip. In other words, if there is a trading activity that causes a change in assets, liabilities or capital, you can enter the information in natural language. It is possible to reduce the errors that can occur in the accounting data by mistake. When the present invention is developed as a computer program and combined with the accounting system, it is expected that the tax agent / accounting office, which acts as a book agent, the business department of a large company, as well as the accounting slips of small and medium-sized companies can be greatly effected. It is possible to apply the automatic journal entry program of the present invention to the previously developed accounting system, so it can be applied to all domestic companies. Especially in the case of small business or self-employment, it is easy to introduce accounting program, so it is easy to grasp the financial situation of the company.

Claims (9)

(a) receiving transaction history information in a natural language sentence; (b) giving a sentence of the transaction history and analyzing the parts of speech including an object and a verb; (c) extracting a main activity factor based on the verb of the analyzed sentence; (d) extracting basic attributes of the main activity factor according to the purpose or contents of the extracted main activity factor; (e) extracting an account subject according to basic attributes of the main activity factor; (f) generating information of necessary items for each account subject based on the transaction history information; And (g) generating an accounting document information based on the extracted information of the account subject and the necessary items. The method of claim 1, wherein step (c) (c1) registering a main activity factor that distinguishes the activity of the transaction; (c2) an automatic journal entry method comprising determining a main activity factor based on the registered main activity factor and the verb of the analyzed phrase. The method of claim 2, Step (c1) registers the main activity factor including priority information of the main activity factor, (c3) if the main activity factors extracted in step (c2) are two or more, further comprising the step of extracting the main activity factors of the input transaction based on the registered priority information; Journal entry method. The method of claim 2, Step (c1) is registered including a questionnaire for selecting the main activity factor in relation to the registered main activity factor, (c4) if there is no main activity factor determined in step (c2), providing the question item to the user, and receiving an answer to the question item, further comprising determining a main activity factor based on the answer Automatic journal method, characterized in that. The method of claim 1, (h) Determining whether the sum of the debit amounts and the sum of the credit amounts in the generated accounting documents is equal, whether the expense account has been debited only and the revenue account has been made only in the credit, so that there is an error in the generated accounting document. Automatic journal method comprising the step of determining the accounting document information by checking. An input unit for receiving natural language sentences related to transaction details and necessary information during journal entry processing of an automatic journal unit; A syntax analysis unit for analyzing the natural language sentence for each component; The main activity factor and the basic attribute information of the main activity factor are extracted according to the components of the analyzed natural language sentence, and the debit and credit account subjects are extracted according to the contents of the main activity factor and the basic attribute information to journal the transaction. Automatic journal to process and confirm the accounting document; And And an output unit for providing the user with information generated during the journal entry processing of the automatic journal unit and providing the determined accounting information to the user. The method of claim 6, wherein the automatic journal unit A main activity factor extracting unit extracting a main activity factor based on the verb of the analyzed natural language sentence, and extracting basic attributes of the main activity factor according to the purpose or content of the extracted main activity factor; Automatic journaling device comprising an accounting subject extraction unit for extracting the debit and credit accounting subject according to the basic attribute of the main activity factor. The method of claim 7, wherein the automatic journal unit It is determined whether the accounting document information generated by processing the transaction history is equal to the sum of the debit amount and the sum of the credit amount of the generated accounting document information, whether the expense account is the debit only and the profit account is the credit only. And checking whether there is an error in the generated accounting document, and determining an accounting document only when there is no error in the inspection. A computer-readable recording medium having recorded thereon a program for executing the method of claim 1 on a computer.