A method and system for intelligently correcting accounting entries

Abstract

The embodiment of the present invention discloses a method and system for intelligently correcting accounting entries, the method comprising: decomposing each entry in a standard answer into scoring points; wherein each scoring point is assigned a corresponding score, and the entries that can be combined are combined, and the entries that can be divided are divided to form an entry combination and a scoring point package; extracting and parsing each entry fragment in the student's answer to form an entry list of the student's answer; dividing the scoring point package, and arranging and combining the scoring points to form a plurality of standard answers; and then comparing and matching each entry in the entry list with the formed plurality of standard answers respectively to realize intelligent correction of accounting entry questions; the beneficial effect is: by arranging and combining the scoring points to form a plurality of standard answers, it is adapted to the different answering situations of the students, and then the CPA accounting entries can be corrected intelligently, which not only improves the correction efficiency, but also reduces the cost of manual correction.

Description

Technical Field

[0001] The present invention relates to the field of computer online education assistance technology, and in particular to an accounting entry intelligent correction method and system. Background Art

[0002] In the CPA accounting exam, there are questions that require students to write answers to various accounting entries. Entries contain three elements: direction, subject, and amount. The direction is divided into debit and credit. Since the number of entries for various accounting businesses is uncertain, the order of multiple entries is also uncertain, and the standard answer is not unique. Students can write simple entries or combine simple entries into a complex entry. The above problems have made it very difficult to intelligently batch accounting entry questions, resulting in the current lack of an intelligent grading system for accounting entry questions in the online education field. However, the use of manual grading has the disadvantages of low grading efficiency and high grading costs. Summary of the invention

[0003] In view of the technical defects in the prior art, the purpose of the embodiments of the present invention is to provide an accounting entry intelligent correction method and system for realizing intelligent correction of accounting entry questions in accounting subject examinations.

[0004] To achieve the above objectives, in a first aspect, an embodiment of the present invention provides an accounting entry intelligent correction method, the method comprising:

[0005] Decompose each entry in the standard answer into scoring points; each scoring point is assigned a corresponding score, and the entries that can be combined are combined, and the entries that can be divided are divided to form entry combinations and scoring point packages;

[0006] Extract and parse each entry fragment in the student's answer to form an entry list of the student's answer;

[0007] Splitting the scoring point package and arranging and combining the scoring points to form a plurality of standard answers;

[0008] Each entry in the entry list is then compared and matched with the multiple standard answers to achieve intelligent correction of accounting entry questions.

[0009] Preferably, the method further comprises:

[0010] Display and proofread the results of intelligent grading;

[0011] Enter the new answer and update it into the standard answer.

[0012] Preferably, the scoring point package is a list composed of multiple entry combinations, and these multiple entry combinations can be replaced with each other; each entry combination includes one or more entries.

[0013] Preferably, the process of forming the entry list is as follows:

[0014] First, extract the entry segments in the student's answer, then perform regular matching, and intercept the student's answer from the position where the first "Debit" appears. Then traverse each line of the student's answer to find out which lines the "Debit" and "Credit" in the answer appear on, thereby forming two lists, one is the direction list; the other is the index list, which corresponds to the line number where each direction in the direction list appears in the student's answer;

[0015] Then, according to the direction list and the index list, calculate the number of entries in the student's answer to form the entry list.

[0016] Preferably, during the intelligent marking, the Jonker-Volgenant algorithm is used to solve the optimal marking match between each entry in the entry list and each scoring point in the standard answer list, so as to calculate the highest score of the student's answer as the student's score.

[0017] In a second aspect, an embodiment of the present invention further provides an intelligent marking system for accounting entries, the system includes:

[0018] A standard answer entry module, which is used to decompose each entry in the standard answer into scoring points; each scoring point is assigned a corresponding score value, and the entries that can be merged are merged, and the entries that can be split are split to form entry combinations and scoring point packages;

[0019] A student answer parsing and extraction module, which is used to extract and parse each entry segment in the student's answer to form an entry list of the student's answer;

[0020] An intelligent marking module, which is used for:

[0021] Split the scoring point package, and perform permutations and combinations on the scoring points to form multiple standard answers;

[0022] Then, compare and match each entry in the entry list with the multiple standard answers formed respectively to achieve intelligent marking of accounting entry questions.

[0023] Preferably, the system further includes a proofreading module, and the proofreading module is used for:

[0024] Display and proofread the results of intelligent marking;

[0025] Enter and update the new answer into the standard answer.

[0026] Preferably, the scoring point package is a list composed of multiple entry combinations, and these multiple entry combinations can be replaced with each other; wherein, each entry combination includes one or more entries.

[0027] Preferably, the process of forming the entry list is as follows:

[0028] First, extract the entry segments in the student's answer, then perform regular matching, and intercept the student's answer from the position where the first "Debit" appears. Then traverse each line of the student's answer to find out which lines the "Debit" and "Credit" in the answer appear in, thereby forming two lists, one is the direction list; the other is the index list, which corresponds to which line each direction in the direction list appears in the student's answer;

[0029] Then, according to the direction list and the index list, calculate the number of entries in the student's answer to form the entry list.

[0030] Preferably, during the intelligent marking, the Jonker-Volgenant algorithm is used to solve the optimal marking matching between each entry in the entry list and each scoring point in the standard answer list, so as to calculate the highest score of the student's answer as the student's score.

[0031] Implementing the embodiments of the present invention, by arranging and combining the scoring points to form multiple standard answers, it can thus adapt to different answering situations of students, and further can intelligently mark CPA accounting entries. On the one hand, it improves the marking efficiency, reduces the cost of manual marking, and can be continuously calibrated and updated, making the marking accuracy relatively high; on the other hand, it can give students timely feedback on the scoring points of their answers, allowing students to correct in time and improving the learning efficiency of students. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for use in the description of the specific embodiments or the prior art.

[0033] Figure 1 is a flowchart of an intelligent marking method for accounting entries provided by an embodiment of the present invention;

[0034] Figure 2 is a flowchart of extracting and analyzing entries from a student's answer provided by an embodiment of the present invention;

[0035] Figure 3 is a structural block diagram of an intelligent marking system for accounting entries provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0036] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0037] Throughout the specification, references to "one embodiment," "an embodiment," "an example," or "an example" mean that a particular feature, structure, or characteristic described in conjunction with the embodiment or example is included in at least one embodiment of the present invention. Therefore, the phrases "in one embodiment," "in an embodiment," "an example," or "an example" appearing in various places throughout the specification do not necessarily all refer to the same embodiment or example. Furthermore, particular features, structures, or characteristics may be combined in one or more embodiments or examples in any appropriate combinations and / or subcombinations.

[0038] It should be noted that, unless otherwise specified, the technical terms or scientific terms used in this application should have the common meanings understood by those skilled in the art to which the invention belongs.

[0039] First, please refer to Figure 1 , an embodiment of the present invention provides an intelligent accounting entry correction method, the method comprising:

[0040] S101, decompose each entry in the standard answer into scoring points; wherein each scoring point is assigned a corresponding score, and mergeable entries are merged, and divisible entries are divisible, so as to form entry combinations and scoring point packages.

[0041] When applied, the scoring point package is a list of multiple entry combinations, and the entry combinations within the scoring point package can be replaced with each other; each entry combination includes one or more entries, and each entry has a corresponding account, amount and score; specifically including:

[0042] First, the standard answer corresponding to the entry question needs to be decomposed into scoring points, each of which has a corresponding entry and score. Because there are different ways to write the same entry, and there are also cases where multiple entries are written together, a scoring point package is constructed to allow a combination of multiple entries to be replaced with another combination of multiple entries. For example, the standard answer is converted into a standard answer list C = [scoring point 1, scoring point 2, scoring point package 1];

[0043] Among them, scoring point package 1 is a list composed of multiple entry combinations. These multiple entry combinations can be replaced with each other, and the student's answer to any one of them is considered correct. For example, scoring point package 1 = [entry combination 1, entry combination 2], and each entry combination includes one or more entries.

[0044] S102, extracting and parsing each entry fragment in the student's answer to form an entry list of the student's answer.

[0045] Specifically, since the format of students answering journal entry questions is relatively fixed, usually writing the debit first and then the credit, a regular matching method is used to extract each entry in the student's answer, including the subject and amount of the entry.

[0046] Please refer to Figure 2 , the process of forming the entry list is as follows:

[0047] Since we don’t know how many entries are written in the student’s answer, we first need to extract the entry fragments in the student’s answer, and then perform regular matching. Since one debit and one credit is one entry, we first intercept the student’s answer from the position where the first “debit” appears, and then traverse each line of the student’s answer to find out the line where the “debit” and “credit” in the answer appear, thus forming two lists, one is the direction list S, for example: S = ['borrow', 'credit', 'borrow', 'credit', 'borrow', 'credit'], then S0, S1, S2, S3, S4, S5 are 'borrow', 'credit', 'borrow', 'credit', 'borrow', 'credit', 6 values ​​respectively, and there are 3 entries in total. The other is the index list I, corresponding to the line where each direction in the direction list appears in the student’s answer, for example: I = [0, 1, 3, 5, 7, 11].

[0048] At this time, based on the direction list and index list, the number of entries answered by the students can be calculated.

[0049] The specific algorithm is as follows:

[0050] ①. Traverse the direction list S and get direction S i , S i Represents the i-th value in the direction list S, where 0<=i<=the length of the direction list S-1, and i represents the subscript of the direction list S.

[0051] ②If S i If it is "borrow", go to step ③, otherwise go back to step ①.

[0052] ③If the direction S i The next direction S j If it is “loan”, go to step ④, otherwise go back to step ①.

[0053] ④. Find direction S according to index list Ii and S j The number of rows that appear is I i , I j , where 0 <= j <= the length of the direction list S - 1, j represents the subscript of the index list I; at this time, determine the starting behavior I of the debit in the entry i , end behavior I j -1; if direction S in the direction list j If it is not the last value in the direction list, then take S j The next value S j+1 , find the number of rows whose occurrence is I according to the index list j+1 , then the starting line number of the credit entry is determined to be I j , the end line number is I j+1 -1, if direction S j is the last value, the starting line number of the credit entry is I j , until the last line of the student's answer;

[0054] Through the above steps, we can find the starting and ending lines of the debit and credit of each entry in the student's answer, so that we can intercept each entry fragment and separate each entry from the student's answer.

[0055] By traversing each line of each entry, using regular matching of Chinese characters and numbers in each line, extracting the account and the corresponding amount respectively. At this time, the debit and credit accounts and amounts of each entry in the student's answer are all extracted for subsequent intelligent correction, and the student's answer entry list A = [Entry 1, Entry 2, Entry 3] is formed.

[0056] S103, splitting the scoring point package and arranging and combining the scoring points to form a plurality of standard answers.

[0057] Specifically, in this embodiment, the above example is used, for example, standard answer list C = [score point 1, score point 2, score point package 1]. Score point package 1 = [entry combination 1, entry combination 2], where entry combination 1 = [score point 3], entry combination 2 = [score point 4, score point 5], entry combination 1 and entry combination 2 are interchangeable, so the standard answer list C can be divided into two score lists: C1 = [score point 1, score point 2, score point 3], C2 = [score point 1, score point 2, score point 4, score point 5]. At this time, the standard answer is divided into multiple standard answers according to the standard answer list C: that is, standard answer lists C1 and C2; in this embodiment, the marked answer has the same meaning as the standard answer list, and multiple standard answers refer to multiple standard answer lists.

[0058] S104, then comparing and matching each entry in the entry list with the multiple standard answers respectively, so as to realize intelligent correction of accounting entry questions.

[0059] Specifically, after the above steps, we get the student's answer entry list A and the standard answer lists C1 and C2; because the order of the entries in the student's answer is unknown, the standard answer for entry 1 in the student's answer is not the score point 1 in the standard answer list, so there is a need for out-of-order correction. Each entry in the student's answer entry list A needs to be corrected with each score point or score point package in the standard answer list C.

[0060] At this time, since there are two standard answer lists, the student's answer A = [entry 1, entry 2, entry 3] needs to be corrected with the two standard answer lists C1 and C2 respectively, and the score with the largest total correction score is selected as the student's score.

[0061] First, use the standard answer list C1 to correct the student's answer list A. When the entry 1 in the student's answer list A is corrected using the scoring point 1, the entry in the scoring point 1 is compared and matched with the entry 1. If the two entries are completely matched in the debit and credit accounts, the student's answer is considered correct and the corresponding scoring point score is given. Similarly, the correction scores of entry 1 and scoring point 2 and scoring point 3 can be calculated. Similarly, the correction scores of entry 2 and entry 3 can be calculated respectively with the scores of scoring point 1, scoring point 2, and scoring point 3. After correction and calculation, the following table can be obtained:

[0062] Score 1 Score 2 Point 3 Entry 1 1.5 2 0 Entry 2 1 3 1 Entry 3 2 1 0.5

[0063] After the entry 1 is corrected with score point 1, the score is 1.5 points, and after the entry 1 is corrected with score point 2, the score is 2. Similarly, after the entry 2 is corrected with score point 1, score point 2, and score point 3, the scores are 1, 3, and 1 respectively, and after the entry 3 is corrected with score point 1, score point 2, and score point 3, the scores are 2, 1, and 0.5 respectively. In the actual correction, since each entry can only be corrected with one score point, if entry 1 corresponds to score point 2, the correction score is 2 points. Since score point 2 has been used to correct entry 1, score point 2 cannot be used to correct entries 2 and 3. At this time, when the standard answer list C1 is used to correct the student's answer, there are 3*2*1=6 correction methods. In order to solve the highest score Score1 among these 6 correction methods, if the exhaustive method is used, it will take a lot of time and the calculation complexity is high. At this time, the problem is converted into a linear assignment problem for solution.

[0064] In order to find the maximum value of the sum of the scores in different rows and columns in the above table as the student's final score score1. At this time, the Jonker-Volgenant algorithm is used to solve the above final score score1. The main principle of the algorithm is:

[0065] First, we need to convert the score table of the above entry list A after correction into a matrix:

[0066]

[0067] Among them, r1, r2, and r3 represent entry 1, entry 2, and entry 3 respectively. c1, c2, and c3 correspond to score point 1, score point 2, and score point 3 respectively.

[0068] ①. Column reduction. If the matrix has n columns, then starting from the nth column to the first column, subtract the maximum value of each element in each column to obtain the following matrix:

[0069]

[0070] The intersection element of the second row r2 and the second column c2 is 0, so r2 and c2 are considered to match. Although the intersection element of the second row r2 and the third column c3 is also 0, since the second row r2 has matched the second column c2, c3 is not matched yet. Similarly, the intersection element of r3 and c1 is 0, so c1 and r3 are matched. The unmatched rows and columns are r1 and c3.

[0071] ②, column inverse reduction,

[0072] Select any matched row and column, such as r3 and c1, select the second largest value -0.5 in r3, and add this value to each element in c1 to obtain the following matrix:

[0073]

[0074] ③. Line contraction

[0075] Following the above steps, subtract the maximum value of the row from each element in the selected row of r3 to obtain the following matrix:

[0076]

[0077] According to this matrix, there is no 0 element in row r1, so r1 is not matched. The first 0 element in row r2 is in column c2, so r2 and c2 match. Similarly, r3 matches c1. At this time, the unmatched rows and columns are r1 and c3 respectively.

[0078] ④. Expand and simplify

[0079] Select the unmatched row r1, subtract the second largest value -1 from each element in the row, and perform inverse column reduction on the c3th column to obtain the following matrix:

[0080]

[0081] At this time, r1 matches c3. Therefore, the final matches are r1 matches c3, r2 matches c2, and r3 matches c1. The scores of the corresponding positions are 0, 3, and 2 respectively. The total score of this student is 2+3+0=5.

[0082] The above steps show that entries 1, 2 and 3 are optimally matched with scoring points 3, 2 and 1 respectively, and each weight of the optimal match is added as the total score score1 of the student after the student's answer list A and the standard answer list C1 are corrected.

[0083] The above process is to use the Jonker-Volgenant algorithm to solve the optimal correction match between each entry in the entry list and each scoring point in the standard answer list, so as to calculate the highest score of the student's answer as the student's score.

[0084] Similarly, the above algorithm can be used to obtain the student's total score score2 after correction of the student's answer list A and the standard answer list C2.

[0085] The student's final score is the maximum of score1 and score2, that is, max(score1, score2);

[0086] Furthermore, in another embodiment, based on the above technical solution, the method further includes:

[0087] Display and proofread the results of intelligent grading;

[0088] Enter the new answer and update it into the standard answer.

[0089] Specifically, after the students' answers have been intelligently graded, the students' answers, standard answers, and grading results will be displayed to the teaching and research staff. If it is found that the students have written the correct answer, but there is no entry in the standard answer, the students' answers need to be converted into scoring points and entered into the standard answer scoring points so that the next time they are encountered, the intelligent grading can correct them correctly. In this way, the grading system will continuously improve the accuracy rate until it reaches a higher value.

[0090] The above technical solution, by arranging and combining the scoring points to form multiple standard answers, can adapt to the different answer situations of students, and then can intelligently correct CPA accounting entries. On the one hand, it improves the correction efficiency, reduces the cost of manual correction, and can be continuously calibrated and updated to achieve a higher correction accuracy; on the other hand, it can provide students with timely feedback on the scoring points of their answers, allowing students to make timely corrections, thereby improving their learning efficiency.

[0091] Based on the same inventive concept, the embodiment of the present invention also provides an accounting entry intelligent correction system, such as Figure 3 As shown, including:

[0092] The standard answer entry module is used to decompose each entry in the standard answer into scoring points; among them, each scoring point is assigned a corresponding score value, and the entries that can be merged are merged, and the entries that can be split are split to form an entry combination and a scoring point package; the scoring point package is a list composed of multiple entry combinations, and these multiple entry combinations can be replaced with each other; among them, each entry combination includes one or more entries.

[0093] The student answer analysis extraction module is used to extract and analyze each entry segment in the student answer to form a list of entries in the student answer; by using the method of regular matching, the direction, subject, and amount of each entry in the student answer are extracted to form a list of entries in the student answer for the next step of marking.

[0094] The process of forming the entry list is as follows:

[0095] First, extract the entry segments in the student answer, then perform regular matching, and intercept the student answer from the position where the first "Debit" appears, and then traverse each line of the student answer to find out which lines the "Debit" and "Credit" in the answer appear in, so as to form two lists, one is the direction list; the other is the index list, which corresponds to the line number where each direction in the direction list appears in the student answer;

[0096] Then, according to the direction list and the index list, calculate the number of entries answered by the student to form the entry list.

[0097] The intelligent marking module is used for:

[0098] Split the scoring point package and perform permutations and combinations of the scoring points to form multiple standard answers;

[0099] Then, compare and match each entry in the entry list with the multiple standard answers formed respectively to achieve intelligent marking of accounting entry questions. When performing intelligent marking, the Jonker-Volgenant algorithm is used to solve the optimal marking match between each entry in the entry list and each scoring point in the standard answer list, and the highest score of the student answer is calculated as the student score.

[0100] Specifically, mark the student answer and each standard answer respectively, and compare whether the direction, subject, and amount of each entry in the student answer match the entry in the standard answer. If they all match, give the corresponding score value, otherwise do not give points. After marking by multiple standard answers, select the highest score of the student as the final score of the student.

[0101] During implementation, to improve the accuracy of marking, the system further includes a proofreading module, and the proofreading module is used for:

[0102] Display and proofread the results of intelligent grading;

[0103] Enter the new answer and update it into the standard answer.

[0104] It should be noted that, for more specific workflows of the intelligent grading system, please refer to the aforementioned method implementation examples, which will not be described in detail here.

[0105] Through the implementation of the above scheme, the problems of how to extract the number and content of entries in students' CPA accounting entry answers and the problem of intelligent grading are solved, and the grading accuracy and efficiency can be continuously improved. The students' scoring points can be fed back in time after they finish the questions, which makes the students' learning more targeted and improves their learning efficiency.

[0106] Those of ordinary skill in the art will appreciate that the modules and steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0107] In the several embodiments provided in this application, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are only schematic, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0108] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or substitutions within the technical scope disclosed by the present invention, and these modifications or substitutions should be included in the protection scope of the present invention.

Claims

1. An intelligent accounting entry correction method, characterized in that: The method includes: Decompose each entry in the standard answer into scoring points; where each scoring point is assigned a corresponding score value, merge the entries that can be combined, and split the entries that can be split to form entry combinations and scoring point packages; the scoring point package is a list composed of multiple entry combinations, and the entry combinations inside this scoring point package can be replaced with each other; where each entry combination includes one or more entries, and each entry has a corresponding subject, amount, and score value; Extract and parse each entry segment in the student's answer to form a list of entries in the student's answer; Split the scoring point package and perform permutations and combinations of the scoring points to form multiple standard answers; Then compare and match each entry in the entry list with the multiple standard answers formed respectively to achieve intelligent grading of accounting entry questions; The process of forming the entry list is as follows: First, extract the entry segments in the student's answer, then perform regular matching, and intercept the student's answer from the position where the first "Debit" appears, and then traverse each line of the student's answer to find out which lines the "Debit" and "Credit" in the answer appear on, so as to form two lists, one is the direction list; the other is the index list, which corresponds to the line number where each direction in the direction list appears in the student's answer; Then, according to the direction list and the index list, calculate the number of entries answered by the student. The specific algorithm is as follows: ①. Traverse the direction list S and get direction S i , S i Represents the i-th value in the direction list S, where 0<=i<=the length of the direction list S-1, and i represents the subscript of the direction list S; ②If S i If it is "borrow", go to step ③, otherwise go back to step ①; ③If the direction S i The next direction S j If it is "loan", go to step ④, otherwise go back to step ①; ④. Find the direction S according to the index list I i and S j The number of rows that appear is I i , I j , where 0 <= j <= the length of the direction list S - 1, j represents the subscript of the index list I; at this time, determine the starting behavior I of the debit in the entry i , end behavior I j -1; if direction S in the direction list j If it is not the last value in the direction list, then take S j The next value S j+1 , find the number of rows whose occurrence is I according to the index list j+1 , then the starting line number of the credit entry is determined to be I j , the end line number is I j+1 -1, if direction S j is the last value, the starting line number of the credit entry is I j , until the last line of the student's answer; Through the above steps, the start line and end line of the debit and credit of each entry in the student's answer are found, so that each entry segment can be intercepted and each entry can be separated from the student's answer; By traversing each line of each entry respectively, use regular expressions to match the Chinese characters and numbers in each line, and extract them as the subject and the corresponding amount respectively; at this time, the subjects and amounts of the debit and credit of each entry in the student's answer are all extracted for subsequent intelligent grading to form the entry list.

2. The method for intelligently correcting accounting entries according to claim 1, characterized in that: The method also includes: Display and proofread the results of intelligent grading; Enter and update the new answer into the standard answer.

3. The intelligent accounting entry correction method according to claim 2 is characterized in that: During the intelligent grading, the Jonker-Volgenant algorithm is used to solve the optimal grading match between each entry in the entry list and each scoring point in the standard answer list to calculate the highest score of the student's answer as the student's score.

4. An intelligent accounting entry correction system, characterized in that the system It includes: A standard answer entry module, which is used to decompose each entry in the standard answer into scoring points; where each scoring point is assigned a corresponding score value, merge the entries that can be combined, and split the entries that can be split to form entry combinations and scoring point packages; the scoring point package is a list composed of multiple entry combinations, and the entry combinations inside this scoring point package can be replaced with each other; where each entry combination includes one or more entries, and each entry has a corresponding subject, amount, and score value; A student answer parsing and extraction module, which is used to extract and parse each entry segment in the student's answer to form a list of entries in the student's answer; An intelligent grading module, which is used for: Split the scoring point package and perform permutations and combinations of the scoring points to form multiple standard answers; Then, each entry in the entry list is compared and matched with the multiple standard answers formed respectively to achieve intelligent marking of accounting entry questions; The process of forming the entry list is as follows: First, extract the entry fragments in the student's answer, then perform regular matching, and intercept the student's answer from the position where the first "Debit" appears. Then traverse each line of the student's answer to find out which lines the "Debit" and "Credit" in the answer appear on, thus forming two lists, one is the direction list; the other is the index list, which corresponds to the line number on which each direction in the direction list appears in the student's answer; Then, according to the direction list and the index list, calculate the number of entries answered by the student. The specific algorithm is as follows: ①. Traverse the direction list S and get direction S i , S i Represents the i-th value in the direction list S, where 0<=i<=the length of the direction list S-1, and i represents the subscript of the direction list S; ②If S i If it is "borrow", go to step ③, otherwise go back to step ①; ③If the direction S i The next direction S j If it is "loan", go to step ④, otherwise go back to step ①; ④. Find the direction S according to the index list I i and S j The number of rows that appear is I i , I j , where 0 <= j <= the length of the direction list S - 1, j represents the subscript of the index list I; at this time, determine the starting behavior I of the debit in the entry i , end behavior I j -1; if direction S in the direction list j If it is not the last value in the direction list, then take S j The next value S j+1 , find the number of rows whose occurrence is I according to the index list j+1 , then the starting line number of the credit entry is determined to be I j , the end line number is I j+1 -1, if direction S j is the last value, the starting line number of the credit entry is I j , until the last line of the student's answer; Through the above steps, the start line and end line of the debit and credit of each entry in the student's answer are found, so that each entry fragment can be intercepted and each entry can be separated from the student's answer; By traversing each line of each entry respectively, regular matching is used to match the Chinese characters and numbers in each line, and they are respectively extracted as the subject and the corresponding amount; at this time, the subjects and amounts of the debit and credit of each entry in the student's answer are all extracted for subsequent intelligent marking to form the entry list.

5. The intelligent accounting entry correction system according to claim 4 is characterized in that: The system further includes a proofreading module, and the proofreading module is used for: Display and proofread the results of intelligent marking; Enter and update the new answer into the standard answer.

6. The intelligent accounting entry correction system according to claim 5 is characterized in that: During the intelligent marking, the Jonker-Volgenant algorithm is used to solve the optimal marking match between each entry in the entry list and each scoring point in the standard answer list, so as to calculate the highest score of the student's answer as the student's score.

Images