A method, apparatus, device, and storage medium for automatically updating financial professional report form data.

Abstract

This invention discloses a method, apparatus, device, and storage medium for automatically updating financial professional report table data. The invention calculates the similarity between a single old table and all new tables in the same major chapter mapping pair. Based on the calculated similarity value, it filters out new tables that meet the requirements according to the most similarity matching rule. Then, it determines whether the filtered new table has already been matched by other old tables. If so, it performs a replacement judgment on the new table according to the matching priority rule, thereby obtaining a new table that establishes a final replacement association with the current old table. This method can perform association matching on all old tables in an old document and uniformly replace the content of the old and new tables. This method has strong generalization capabilities and stable ability to replace the table content of financial institution documents.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and in particular to a method, apparatus, device, and storage medium for automatically updating financial professional report form data. Background Technology

[0002] Updating financial institution filings refers to revising, updating, or modifying information in these filings to reflect the latest operational status, financial condition, risk factors, and other relevant information. This is a fundamental business practice to ensure potential investors have access to up-to-date and accurate information to make informed investment decisions.

[0003] Updating documents in financial institutions is often a tedious, repetitive, and time-consuming task. The updated content is often extensive but consists of formulaic content replacements, such as replacing sentences and tables with specified formats and updating data, dates, amounts, and percentages. These sentences and tables are semantically similar but have unpredictable substantive differences. Therefore, a system is needed that can use rules to determine text similarity and automatically replace document content, thereby eliminating repetitive manual tasks and improving office efficiency.

[0004] Currently, the commonly used method is the "fill-in-the-blank" approach, which establishes a one-to-one mapping between the content of the file to be modified and the content of the template file, and updates the old file content by updating the new template. However, this method has limitations. For each type of file being updated, a corresponding template file for that type of file needs to be created, and replacement judgment logic and mapping relationships need to be established. This method has a large limitation on the content that can be replaced, cannot flexibly adjust the replacement range, and the template lacks generalization.

[0005] Therefore, a new technology is needed that can generalize file templates and use large models to calculate text similarity for file content replacement. Summary of the Invention

[0006] In view of this, the present invention provides a method, apparatus, device and storage medium for automatically updating financial professional report form data. The present invention has strong generalizability, can replace the content of financial institution document forms and has stable capabilities.

[0007] A method for automatically updating data in financial professional report forms includes the following steps:

[0008] S1 extracts the table titles, the chapters to which the tables belong, and the table content from the new and old files, respectively. The new file refers to the template file, and the old file refers to the file to be modified.

[0009] S2, normalize all extracted table titles, table chapters, and table contents;

[0010] S3, using chapters as the mapping scale, obtains the chapter mapping dictionary between new and old files;

[0011] S4. Based on the chapter mapping dictionary, calculate the similarity between a single old table and all the remaining new tables after removing all new tables that have been associated with the old table (and failed to replace the original associated old table).

[0012] S5. Based on the calculated similarity value between the new and old tables, filter and match the new table that meets the conditions according to the set most similarity matching rule. If no new table that meets the most similarity matching rule is found, the current old table is considered to have failed to match and will not be replaced. Repeat step S4 to match the next old table. If a new table that meets the most similarity matching rule is matched, proceed to step S6.

[0013] S6, establish a preliminary replacement association between the new table that satisfies the most similar matching rule and the current old table, and then determine whether the new table that has established a preliminary replacement association is a new table that has been matched by other old tables. If so, execute S7; otherwise, the new table that has established a preliminary replacement association is the new table that has established a final replacement association with the current old table.

[0014] S7: Based on the set matching priority rules, the new table with the initial replacement association is evaluated for replacement. If the replacement is successful, the new table with the initial replacement association becomes the new table with the final replacement association with the old table, and the old table that was replaced is re-executed from S4 to S7. If the replacement fails, the old table is re-executed from S4 to S7.

[0015] S8. Repeat steps S4-S7 to match the next old table until all old tables have been matched.

[0016] S9 replaces the content of all new and old tables that have implemented the replacement association.

[0017] Preferably, the similarity between the old and new tables includes a title similarity S. title Table content similarity S cell And the weighted score S for similarity between table titles and content. title_cell .

[0018] Preferably, the specific steps in step S4 for calculating the similarity between a single old table and all the remaining new tables after removing all new tables that were previously associated with the old table (and whose replacement of the original associated old table failed) within the same major chapter mapping pair are as follows:

[0019] S41, retrieve the title body keywords of the current old table, and match new tables with the same title body keywords from all new tables belonging to the same major chapter mapping pair. Old tables with titles that do not contain any body keywords can only be matched with new tables with titles that do not contain any body keywords. The matched new tables are called similar new tables.

[0020] S42, calculate the title similarity S between the current old table and each similar new table. title Similarity S to table content cell ;

[0021] S43, Title similarity S between the current old table and each similar new table title Similarity S to table content cell Calculate the weighted score S for the similarity of the corresponding table titles and content. title_cell ;

[0022] S44, calculate all the title similarity values ​​S title Table content similarity value S cell The weighted score S for similarity between table title and content title_cell Sort the titles according to their scores from highest to lowest to obtain a list of titles with similarity scores {S}. title}, Table content similarity sorting list {S cell} and a list sorted by weighted score of table title and content similarity {S title_cell}

[0023] Preferably, in step S42, the title similarity S between the current old table and a single similar new table is calculated. title The specific steps are as follows:

[0024] By referring to the set subject keyword list, extract subject keywords from the current old table and similar new tables respectively;

[0025] The title matching score S is determined based on the similarity between the subject keywords in the current old table and the subject keywords in similar new tables. key ;

[0026] Obtain the semantic vectors of the table titles of the current old table and similar new tables respectively;

[0027] The cosine similarity S between the old and new title vectors is calculated based on the semantic vectors of the old and new title vectors. vec Cosine similarity S between old and new title vectors vec As a reference value for title similarity;

[0028] Based on the title matching score S keyCosine similarity S between the old and new title vectors vec Calculate the title similarity S between the current old table and similar new tables. title .

[0029] Preferably, if the subject keywords of the current old table are exactly the same as the subject keywords of a similar new table, then the title matching score S is [value missing]. key =1;

[0030] If the subject keywords in the current old table match the subject keywords in a similar new table, the title matching score is S. key =0.5;

[0031] If the subject keywords in the current old table do not match the subject keywords in a similar new table, the title matching score will be S. key =0.

[0032] Preferably, the cosine similarity S between the new and old title vectors vec The calculation formula is:

[0033]

[0034] Among them, v old v is the semantic vector for the title of the current old table. new For the title semantic vector of a similar new table;

[0035] The title similarity S between the current old table and similar new tables title The calculation formula is:

[0036]

[0037] Among them, w k and w t All of these are set weight values.

[0038] Preferably, if the current old table has no header, then the header similarity S between the current old table and similar new tables is calculated. title Set to 0.

[0039] Preferably, in step S42, the table content similarity S between the current old table and a single similar new table is calculated. cell The specific steps are as follows:

[0040] Calculate the initial similarity S between the contents of the old table (after normalization) and the new table (after normalization). cell_0 ;

[0041]

[0042] Where, {O1,…,On} represents the set of all cell contents of the current old table after normalization, {N1,…,N n} represents the collection of all cell contents after normalization of the similar new table, and LENGTH() returns the length of the collection;

[0043] Based on the relationship between the smallest section above the current old table and the smallest section above the similar new table, the initial similarity S of the table content is calculated. cell_0 Add the minimum chapter relationship weight to the table to obtain the table content similarity S between the current old table and similar new tables. cell .

[0044] Preferably, if the minimum chapter text of both the current old table and the similar new table is not empty and the minimum chapter text of the two tables is contained within each other, if S cell_0 ≥T contain_chapter Then S cell =S cell_0 +W contain_chapter ;

[0045] If the minimum chapter text of both the current old table and the similar new table is not empty and the minimum chapter text of the two tables is equal, then S cell_0 =0, then S cell =W contain_chapter ;

[0046] If the minimum chapter text of both the current old table and the similar new table is not empty and the minimum chapter text of the two tables is equal, then S cell_0 If S ≠ 0, then S cell =S cell_0 +W same_chapter ;

[0047] Among them, T contain_chapter To set the table content similarity threshold, W contain_chapter and W same_chapter These are all set minimum chapter relationship weight values ​​for the tables.

[0048] Preferably, if the difference between the smallest chapter text of the current old table and the similar new table is a specified keyword, S cell =-1.

[0049] Preferably, the specific steps in step S5 for filtering new tables that meet the conditions based on the calculated similarity values ​​of the old and new tables according to the set most similarity matching rules are as follows:

[0050] S51, Determine the title similarity ranking list {S title S, the highest score in} title_max Does it satisfy S? title_max >Ttitle_top If the condition is met, the new table corresponding to the title similarity value will be replaced by the current old table; otherwise, step S52 will be executed.

[0051] S52, Determine the title similarity ranking list {S title S, the highest score in} title_max Does it satisfy 0? title_max <T title_min If the condition is met, the current old table is considered to have failed to match and will not be replaced. Step S4 is repeated to match the next old table. Otherwise, step S53 is executed.

[0052] S53, Determine the weighted score ranking list of table titles and content similarity {S title_cell S, the highest score in} title_cell_max Does it satisfy S? title_cel_maxl >T title_cell If the conditions are met, a replacement association is established between the new similar table corresponding to the weighted score of the similarity between the table title and content and the current old table; otherwise, step S54 is executed.

[0053] S54, Sort the table content similarity list {S cell S, the highest score in} cell_max Does it satisfy S? cell_max <T cell_min If the condition is met, the current old table is considered to have failed to match and will not be replaced. Step S4 is repeated to match the next old table; otherwise, step S55 is executed.

[0054] S55, Sort the table content similarity list {S cell S, the highest score in} cell_max Does it satisfy S? cell_max >T cell If the condition is met, a replacement association is established between the new similar table corresponding to the table content similarity value and the current old table; otherwise, step S56 is executed.

[0055] S56, Determine the title similarity ranking list {S title S, the highest score in} title_max Does it satisfy S? title_max >T title If the conditions are met, the new table corresponding to the title similarity value will be linked to the current old table for replacement; otherwise, the current old table will be considered to have failed to match and will not be replaced. This step S4 will be repeated to match the next old table.

[0056] Among them, T title_top T title_min and T title All are set title similarity thresholds, T title_cell ​T is a weighted threshold for the similarity of table titles and content. cell_min and T cell These are all set table content similarity thresholds.

[0057] Preferably, the T title_min <T title <T title_top T cell_min <T cell .

[0058] Preferably, the title similarity threshold T title_min =0.5, T title =0.829, T title_top =0.98, Table content similarity threshold T cell_min =0.04, T cell =0.082, weighted threshold T for table title and content similarity title_cell =0.7.

[0059] Preferably, the specific steps in step S7 of performing a replacement judgment on the new table with which the initial replacement association has been established according to the set matching priority rules, and determining the new table with which the final replacement association has been established with the current old table based on the replacement judgment result, are as follows:

[0060] S71, let the current old table be the first old table, and let the other old tables corresponding to the new table that has been matched by other old tables be the second old table;

[0061] Obtain the final matching state of the first and second old tables when applying the most similarity matching rule, and obtain the final matching similarity score S of the first old table. replace The final matching similarity score S of the _new and second old tables replace _old;

[0062] S72, determine the priority status code C of the first old table according to the priority status code determination rules. re _new and the priority status code C of the second old table re _old;

[0063] S73, determine the matching priority of the first old table and the second old table based on the final matching similarity score and priority status code of the first old table and the second old table. If the matching priority of the first old table is higher than that of the second old table, take the new table that has established a preliminary replacement association as the new table that has established a final replacement association with the first old table, and cancel the replacement association between the new table and the second old table and repeat step S4 until the second old table matches the new table or does not match the new table.

[0064] If the matching priority of the first old table is lower than that of the second old table, then cancel the replacement association between the first old table and the new table with which the initial replacement association has been established, keep the replacement association between the second old table and the new table unchanged, and execute step S4 until the first old table matches the new table or does not match the new table.

[0065] If the first or second old table does not match the new table, it is considered a match failure and will not be replaced.

[0066] Preferably, the priority status code determination rule is as follows:

[0067] If the old table achieves a high title similarity score when applying the most similarity matching rule, thus satisfying the requirement to establish a replacement association, then C re =3;

[0068] If the final matching state of the old table when performing the most similarity matching rule is such that the weighted similarity score of the table title and content meets the requirements for establishing a replacement association, then C re =2;

[0069] If the final matching state of the old table when performing the most similarity matching rule is such that the table content similarity meets the requirements for establishing a replacement association, then C re =1;

[0070] If the final matching state of the old table when performing the most similarity matching rule is that the table title similarity meets the requirements for establishing a replacement association, then C re =0;

[0071] If the old table fails to match when performing the most similarity matching rule, then C re =-1.

[0072] Preferably, the specific steps in step S73 for determining the matching priority between the first old table and the second old table based on the final matching similarity score and priority status code of the first old table and the second old table are as follows:

[0073] Determine whether the final matching similarity score and priority status code of the first and second old tables meet condition C. re _new>C re _old, or C re _new=C re _old and S replace _new>S replace If the condition _old is met, it means that the matching priority of the first old table is higher than that of the second old table; otherwise, the matching priority of the first old table is lower than that of the second old table.

[0074] Preferably, the final matching similarity scores of the first old table and the second old table are the similarity scores corresponding to the final matching state they reached when performing the most similar matching rule.

[0075] An apparatus for automatically updating financial professional report form data using the method described above, comprising:

[0076] The file processing module is used to extract table titles, the chapters to which tables belong, and table content from new and old files, and to normalize all extracted table titles, the chapters to which tables belong, and table content.

[0077] The document chapter mapping module is used to obtain a chapter mapping dictionary between new and old files, using chapters as the mapping scale;

[0078] The new and old table matching module is used to calculate the new and old table similarity between a single old table and all the remaining new tables after removing all new tables that have been associated with the old table (and failed to replace the original associated old table) in the same major chapter mapping pair according to the chapter mapping dictionary. Based on the calculated new and old table similarity value, a new table that meets the most similar matching rule is matched. After establishing a preliminary replacement association between the new table that meets the most similar matching rule and the current old table, it is determined whether the new table that has established a preliminary replacement association is a new table that has been matched by other old tables. If so, the replacement judgment is performed on the new table that has established a preliminary replacement association according to the set matching priority rule.

[0079] The module for replacing content in old and new tables is used to uniformly replace the content in all old and new tables that have been linked for replacement.

[0080] A computer device includes a processor and a memory, wherein the memory stores at least one program, which is loaded and executed by the processor to implement the method described above.

[0081] A computer-readable storage medium storing at least one program that is loaded and executed by a processor to implement the method described above.

[0082] A computer program product comprising at least one program segment stored in a computer-readable storage medium; a processor of a communication device reads the at least one program segment from the computer-readable storage medium, and the processor executes the at least one program segment to cause the communication device to perform the method described above.

[0083] The beneficial effects of this invention are:

[0084] This invention calculates the similarity between a single old table and all new tables in the same major chapter mapping pair. Based on the calculated similarity value, it filters out new tables that meet the requirements according to the most similarity matching rule. Then, it determines whether the filtered new table has already been matched by other old tables. If so, it replaces the new table according to the matching priority rule, thereby obtaining a new table that establishes a final replacement association with the current old table. This method can be used to match and associate all old tables in an old file and uniformly replace the content of the old and new tables. This method has strong generalization ability and stable capability for replacing the table content in financial institution documents. Attached Figure Description

[0085] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0086] Figure 1 This is a flowchart illustrating the automatic data update process for financial professional report tables in this invention.

[0087] Figure 2 It is a matching subprocess for a single old table. Detailed Implementation

[0088] To make the objectives, technical solutions, and advantages of this invention clearer, the invention is described below with reference to specific embodiments shown in the accompanying drawings. However, it should be understood that these descriptions are merely exemplary and not intended to limit the scope of the invention. Furthermore, descriptions of well-known structures and technologies are omitted in the following description to avoid unnecessarily obscuring the concept of the invention.

[0089] The terminology used in this disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. The singular forms “a,” “the,” and “the” as used in this disclosure and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

[0090] It should be understood that although the terms "first" and "second" may be used in this disclosure to describe various information, such information should not be limited to these terms and should not be construed as indicating or implying relative importance. These terms are used only to distinguish information of the same type from one another. For example, without departing from the scope of this disclosure, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0091] Updating financial institution documents refers to revising, updating, or modifying information within these documents to reflect the latest operational status, financial condition, and risk factors. The work primarily involves replacing sentences and table content in specified formats, and updating data, dates, amounts, percentages, etc. Existing methods for updating document content through fill-in-the-blank techniques lack flexibility, are not universally applicable to various document types, and may encounter situations where table content is identical but table attributes differ (e.g., two tables with nearly identical content but different titles and sections). This frequently occurs in financial institution documents, hindering similarity assessment. This invention provides an automatic data update method for financial professional report tables. This method possesses strong generalization and versatility, applicable to various types of document updates, replacing table content in financial institution documents, and exhibits stability and robustness.

[0092] The method for automatically updating financial professional report table data provided in this invention includes the following steps:

[0093] S1 extracts the table titles, the chapters to which the tables belong, and the table content from the new and old files, respectively. The new file refers to the template file, and the old file refers to the file to be modified.

[0094] S2, normalize all extracted table titles, table chapters, and table contents;

[0095] S3, using chapters as the mapping scale, obtains the chapter mapping dictionary between new and old files;

[0096] S4. Based on the chapter mapping dictionary, calculate the similarity between a single old table and all the remaining new tables after removing all new tables that have been associated with the old table (and failed to replace the original associated old table) in the same major chapter mapping pair.

[0097] S5. Based on the calculated similarity value between the new and old tables, filter and match the new table that meets the conditions according to the set most similarity matching rule. If no new table that meets the most similarity matching rule is found, the current old table is considered to have failed to match and will not be replaced. Repeat step S4 to match the next old table. If a new table that meets the most similarity matching rule is matched, proceed to step S6.

[0098] S6, establish a preliminary replacement association between the new table that satisfies the most similar matching rule and the current old table, and then determine whether the new table that has established a preliminary replacement association is a new table that has been matched by other old tables. If so, execute S7; otherwise, the new table that has established a preliminary replacement association is the new table that has established a final replacement association with the current old table.

[0099] S7: Based on the set matching priority rules, the new table with the initial replacement association is evaluated for replacement. If the replacement is successful, the new table with the initial replacement association is the new table with the final replacement association with the old table. The old table that was replaced is re-executed from S4 to S7. If the replacement fails, the old table is re-executed from S4 to S7.

[0100] S8. Repeat steps S4-S7 to match the next old table until all old tables have been matched.

[0101] S9 replaces the content of all new and old tables that have implemented the replacement association.

[0102] To better understand the technical solution of the present invention, the present invention will be described in detail below with reference to the accompanying drawings.

[0103] The method for automatically updating financial professional report table data provided in this invention specifically includes the following steps:

[0104] S1 parses the new and old files respectively, and extracts the table titles, the chapters to which the tables belong, and the table content from the new and old files.

[0105] For table titles: Extract the first line of text located above the table that ends without punctuation (except for "Unit: Ten Thousand Yuan", "Unit: Hundred Million Yuan", or "Unit: Yuan", etc.) and without a number.

[0106] For the chapter to which the table belongs: extract the text content of each level that starts with a number and has a different number and no punctuation at the end, starting from the beginning of the table. The number format is usually: I, 1, (1), ①. Chapters are generally multi-level structures.

[0107] For table content: Extract all text content from all cells in the table.

[0108] S2, normalize all the extracted table titles, the chapters to which the tables belong, and the table contents. For example, modify all time expressions to "during the reporting period", remove redundant punctuation marks, spaces, and illegal characters from all texts.

[0109] Specifically, the normalization process includes the following steps:

[0110] Normalize the table titles: Remove expressions such as "Table:", "Table", "Table 1-1", "1.", etc. at the beginning; remove the character "Table" at the end.

[0111] Normalize the chapters to which the tables belong: Remove the numerical serial numbers starting with one, 1, (1), ①, etc.; remove similar expressions starting with "Chapter 1", "Section 1", etc., and retain the text part of the subsequent chapter. For example, for "Chapter 6 The Issuer's Main Financial Conditions", remove the text "Chapter 1" and the subsequent space, and retain the text "The Issuer's Main Financial Conditions".

[0112] Normalize the table contents: For all the contents in the table except the contents to be hidden, the contents to be hidden are: keyword texts including "Project", "Total", "Amount", "Percentage", "Ratio", "Category", "Subtotal", "Serial Number", etc.

[0113] Use the texts of the normalized table titles, the chapters to which the tables belong, and the table contents to match the new tables in the new file according to the subsequent step rules.

[0114] S3, taking the chapter as the mapping scale, use the GPT-4 large multi-modal model to obtain the chapter mapping dictionary between the new and old files.

[0115] The chapter mapping dictionary refers to the chapter mapping relationship composed of the highest-level chapters to which the tables in the new and old files belong.

[0116] The specific steps to obtain the chapter mapping dictionary between the new and old files are: Input the maximum-level chapter information of the old file and the multi-level chapter information of the new file into the GPT-4 large multi-modal model respectively. The GPT-4 large multi-modal model judges the most similar maximum-level chapter in the new file for each maximum-level chapter in the old file according to the semantics. <​​​​​​The chapter mapping relationship between the old table A and the new table a is "Issuer's Main Financial Condition" corresponding to "Issuer's Main Financial Condition" (normalized major chapter text mapping). The chapter "Chapter 1: Issuer's Main Financial Condition" corresponding to "Chapter 6: Issuer's Main Financial Condition" is the highest-level chapter in the mapping between the old and new tables.

[0120] S4. Based on the chapter mapping dictionary, calculate the similarity between a single old table and all remaining new tables in the same major chapter after removing all old tables that have been associated with the old table (and failed to replace the original associated old tables).

[0121] Specifically, in S41, obtain the title and main keywords of the current old table. Based on the title and main keywords of the current old table, match new tables with the same title and main keywords from all remaining new tables belonging to the same major chapter mapping pair after removing all new tables that have been associated with the old table (and failed to replace the original associated old table). For old tables without title and main keywords, only new tables that also do not have title and main keywords can be matched. The matched new tables are called similar new tables.

[0122] The title subject keyword is used to limit the matching range of new and old tables that both have titles (some tables may not have titles). For example, an old table with an old title containing the subject keyword can only match a new table with an old title containing the subject keyword, and an old table with an old title without any subject keyword can only match a new table with a new title without any subject keyword. For example, an old table with a title containing the subject keyword "parent company" can only match a new table with a title containing the subject keyword "parent company".

[0123] Assuming that the old table A in the old document belongs to the main chapter "Chapter 1: Main Financial Conditions of the Issuer" and the title "Parent Company Balance Sheet of the Issuer as of the End of 2020-2022 and March 2023" contains the main keyword "parent company", then based on the title main keyword "parent company", a new table a is matched from the "Chapter 6: Main Financial Conditions of the Issuer" in the new document. This new table a also has the title "Parent Company Balance Sheet of the Issuer as of the End of 2021-2023 and March 2024" and contains the main keyword "parent company".

[0124] Assuming that the old table A in the old document belongs to the main chapter "Chapter 1: Main Financial Conditions of the Issuer" and the title "Detailed Composition of Period Expenses of the Issuer from the End of 2020 to the End of 2022 and March 2023" does not contain any main keywords, the new document's "Chapter 6: Main Financial Conditions of the Issuer" will match a new table a with the same title "Detailed Composition of Period Expenses of the Issuer from the End of 2021 to the End of 2023 and March 2024" which also does not contain any main keywords.

[0125] Suppose that there are three new tables a, b, and c that match the old table A in the old document. These three matching new tables a, b, and c are called similar new tables.

[0126] S42, calculate the title similarity S between the current old table and each similar new table. title :

[0127] (1) According to the set subject keyword table, extract the subject keywords from the titles of the current old table and similar new tables respectively. For example, the title "Profitability Structure Table of Issuer at the End of 2021-2023 and March 2024" contains the subject keyword "profitability".

[0128] A table title may contain zero or more subject keywords;

[0129] (2) Determine the title matching score S based on the similarity between the subject keywords in the current old table and the subject keywords in similar new tables. key (Also known as subject keyword hit weight score), that is, subject keywords are extracted from both the old and new titles, and the title matching score is increased based on the subject keyword hit rate;

[0130] If the subject keywords in the current old table are exactly the same as the subject keywords in a similar new table, then the title matching score is S. key =1, for example, if the subject keywords of the current old table and the similar new table are both "cash flow", "profitability", and "operating efficiency", then the subject keywords of the two tables are completely identical, S key =1;

[0131] If the subject keywords in the current old table match the subject keywords in a similar new table, the title matching score is S. key =0.5, if the subject key of the current old table is "cash flow", while the subject key of the similar new table is "cash flow" and "profitability", then the subject key parts of the two are the same, S key =0.5;

[0132] If the subject keywords in the current old table do not match the subject keywords in a similar new table, the title matching score will be S.key =0, if the subject key of the current old table is "cash flow" and the subject key of the similar new table is "profitability", then the subject keys of the two are inconsistent, S key =0.

[0133] (3) Obtain the semantic vectors of the table titles of the current old table and similar new tables respectively through the BGE semantic vector model;

[0134] V new ={v new_1 ,…,v new_n}

[0135] V old ={v old_1 ,…,v old_n}

[0136] V new V represents the set of semantic vectors for table headers of similar new tables. old A set of semantic vectors representing the table headers of the old table.

[0137] (4) Calculate the cosine similarity S between the old and new title vectors based on the semantic vectors of the table titles of the current old table and similar new tables. vec Cosine similarity S between old and new title vectors vec As a reference value for title similarity, a title similarity matrix is ​​obtained;

[0138] Cosine similarity S between old and new title vectors vec The calculation formula is:

[0139]

[0140] Among them, v old v is the semantic vector for the title of the current old table. new For the semantic vector of the title of a similar new table.

[0141] (5) Match the score S based on the title key Cosine similarity S between the old and new title vectors vec Calculate the title similarity S between the current old table and similar new tables. title .

[0142] The similarity S between the headers of the current old table and similar new tables title The calculation formula is:

[0143]

[0144] Among them, w k and w t All of these are set weight values.

[0145] w k and w t The specific value is set based on experience; in this embodiment, the optimal setting is w. k =0.4, w t =0.6.

[0146] If the old table has no header, then the header similarity S between it and all similar new tables is used. title Set all values ​​to 0.

[0147] For example, following the calculation method described above, the title similarity value S between the old table A and similar new tables a, b, and c can be calculated respectively. title_Aa S title_Ab S title_Ac .

[0148] Calculate the table content similarity S between the current old table and each similar new table. cell :

[0149] (1) Calculate the initial similarity S between the contents of the old table and the new table after normalization of all cell contents. cell_0 ;

[0150]

[0151] Where, {O1,…,O n} represents the set of all cell contents of the current old table after normalization, {N1,…,N n} represents the collection of all cell contents after normalization of the similar new table, and LENGTH() returns the length of the collection;

[0152] (2) Based on the relationship between the smallest section above the current old table and the smallest section above the similar new table, the initial similarity S of the table content is calculated. cell_0 Add the minimum chapter relationship weight to the table to obtain the table content similarity S between the current old table and similar new tables. cell ,Right now:

[0153] If the smallest section text of both the current old table and the similar new table is non-empty and their smallest section texts are in an inclusive relationship (for example, if the smallest section text of the current old table is "Profitability" and the smallest section text of the similar new table is "Profitability Analysis", then the smallest section texts of both are non-empty and are in an inclusive relationship), if S cell_0 ≥T contain_chapter Then S cell =S cell_0 +W contain_chapter ;

[0154] If the smallest section text of both the current old table and the similar new table is non-empty and their smallest section texts are equal (for example, if the smallest section text of the current old table is "profitability" and the smallest section text of the similar new table is also "profitability", then their smallest section texts are non-empty and equal), if S cell_0 =0, then S cell =W contain_chapter ;

[0155] If the minimum chapter text of both the current old table and the similar new table is not empty and the minimum chapter text of the two tables is equal, then S cell_0 If S ≠ 0, then S cell =S cell_0 +W same_chapter ;

[0156] Among them, T contain_chapter To set the table content similarity threshold, W contain_chapter and W same_chapter All are the minimum chapter relationship weight values ​​set for the tables, T contain_chapter W contain_chapter and W same_chapter The specific value is set based on experience; in this embodiment, the optimal setting is T. contain_chapter =0.019, W contain_chapter =0.31, W same_chapter =0.35.

[0157] Preferably, if the difference between the smallest chapter text of the current old table and the similar new table is a specified keyword, then S is directly set. cell =-1.

[0158] For example, following the calculation method described above, the title similarity value S between the old table A and similar new tables a, b, and c can be calculated respectively. cell_Aa S cell_Ab S cell_Ac .

[0159] S43, Title similarity S between the current old table and each similar new table title Similarity S to table content cell Calculate the weighted score S for the similarity of the corresponding table titles and content. title_cell ;

[0160] Table title and content similarity weighted score S title_cell The calculation formula is:

[0161]

[0162] w tt and w tc w is the weight value. tt and wt c The specific value is set based on experience; in this embodiment, the optimal setting is w. tt =0.6, w tc =0.4.

[0163] For example, the weighted score S of the similarity between the old table A and similar new tables a, b, and c in terms of table titles and content can be calculated separately. title_cell_Aa S title_cell_Ab S title_cell_Ac .

[0164] S44, calculate all the title similarity values ​​S title Table content similarity value S cell The weighted score S for similarity between table title and content title_cell Sort the titles according to their scores from highest to lowest to obtain a list of titles with similarity scores {S}. title}, Table content similarity sorting list {S cell} and a list sorted by weighted score of table title and content similarity {S title_cell}

[0165] For example, the title similarity value S calculated according to step S42 above... title_Aa S title_Ab S title_Ac (Assume S) title_Aa >S title_Ab >S title_Ac Sort the titles according to their scores from highest to lowest to obtain a list of titles with similarity scores {S}. title};

[0166] The title similarity value S calculated according to step S42 above will be used as a reference. cell_Aa S cell_Ab S cell_Ac (Assume S) cell_Ab >S cell_Aa >S cell_Ac Sort the tables according to their scores from highest to lowest to obtain a sorted list of table content similarity. cell};

[0167] The table title and content similarity weighted score S calculated according to step S43 above will be used as the basis for the calculation. title_cell_Aa S title_cell_Ab S title_cell_Ac (Assume S) title_cell_Ac >S title_cell_Aa >S title_cell_Ab)Sort in descending order of scores to obtain a list of sorted table titles and content similarity weighted scores {S title_cell}.

[0168] S5. According to the calculated similarity value between the new and old tables, filter and match new tables that meet the conditions according to the set most similar matching rule. If no new table that meets the most similar matching rule is found, it is considered that the current old table fails to match and is not replaced. Repeat step S4 to match the next old table; if a new table that meets the most similar matching rule is matched, execute step S6.

[0169] The specific steps for filtering and matching new tables that meet the conditions according to the calculated similarity value between the new and old tables according to the set most similar matching rule are as follows:

[0170] S51. Judge whether the largest S in the sorted list of title similarity scores {S title} satisfies S title_max >T title_max >T title_top , if it is satisfied, a preliminary replacement association is established between the similar new table corresponding to the title similarity value and the current old table (for example, if S title has the largest score in {S title_Aa}, judge whether S title_Aa satisfies S title_Aa >T title_top , if it is satisfied, establish a preliminary replacement association between the new table a corresponding to S title_Aa and the current old table), otherwise execute step S52;

[0171] S52. Judge whether the largest S in the sorted list of title similarity scores {S title} satisfies 0 < S title_max < T title_max <T title_min (for example, judge whether S title_Aa satisfies 0 < S title_Aa < T title_min ), if it is satisfied, it is considered that the current old table fails to match and is not replaced. Repeat step S4 to match the next old table, otherwise execute step S53;

[0172] S​​​​​​​​​​Whether S is satisfied title_cell_Ac >T title_cell If it is satisfied, in this step, the new table c corresponding to S title_cell_Ac is initially associated with the current old table for replacement. Otherwise, execute step S54;

[0173] S54. Judge whether the S with the largest score in the sorted list of table content similarity{S cell} cell_max satisfies S cell_max <T cell_min (For example, in{S cell} cell_Ab S has the largest score. Judge whether S cell_Ab satisfies S cell_Ab <T cell_min ). If it is satisfied, it is regarded that the current old table fails to match and no replacement is made. Repeat step S4 to match the next old table. Otherwise, execute step S55;

[0174] S55. Judge whether the S with the largest score in the sorted list of table content similarity{S cell} cell_max satisfies S cell_max >T cell . If it is satisfied, the similar new table corresponding to this table content similarity value is initially associated with the current old table for replacement (for example, in{S[[ID=s39]] cell} cell_Ab S has the largest score. Judge whether S cell_Ab satisfies S cell_Ab >T cell . If it is satisfied, in this step, the new table b corresponding to S cell_Ab is initially associated with the current old table for replacement). Otherwise, execute step S56;

[0175] S56. Judge whether the S with the largest score in the sorted list of title similarity{S title} title_max ]>satisfies S title_max >T title . If it is satisfied, the similar new table corresponding to this title similarity value is initially associated with the current old table for replacement (for example, in{S title} title_Aa S has the largest score. Judge whether S title_Aa satisfies S title_Aa >T title . If it is satisfied, in this step, the new table a corresponding to S title_Aa is initially associated with the current old table for replacement). Otherwise, it is regarded that the current old table fails to match and no replacement is made. Repeat this step S4 to match the next old table;

[0176] Among them, T title_topT title_min and T title All are set title similarity thresholds, T title_cell T is a weighted threshold for the similarity of table titles and content. cell_min and T cell All of these are set table content similarity thresholds, and T title_min <T title <T title_top T cell_min <T cell .

[0177] T title_top T title_min 、、T title_cell T cell_min and T cell The specific values ​​are all set based on experience. In this embodiment, T title_to =0.98, T title_cell =0.7, T cell =0.082, T title =0.829, T cell_min =0.04, T title_min =0.5.

[0178] Preferably, if the text difference between the smallest chapter to which the new and old tables belong is a specified keyword, then no replacement will be made.

[0179] Preferably, if the specified keyword pairs appear in both the old and new table titles, they will not be replaced.

[0180] Taking the above example again, when the old table A is matched with similar new tables a, b, and c according to the set most similarity matching rule, if the procedure terminates after executing step S51, then the new table a that meets the criteria of this step will establish a preliminary replacement association with the current old table; if the criteria of S51 and S52 are not met, the procedure will not terminate until step S53, then the new table c that meets the criteria of S53 will establish a preliminary replacement association with the current old table; if the criteria of S51-S55 are not met, the procedure will not terminate until the final step S56, then the new table a that meets the criteria of S56 will establish a preliminary replacement association with the current old table.

[0181] S6, establish a preliminary replacement association between the new table that satisfies the most similar matching rule and the current old table (taking the above example as an example, assuming that in step S5 the old table A and the similar new tables a, b, c are executed according to the set most similar matching rule until the program terminates in S53, then establish a preliminary replacement association between the new table c and the current old table).

[0182] Then determine whether the new table (new table c) that has established a preliminary replacement association is a new table that has been matched by other old tables (i.e., whether there is an old table B that has established a preliminary replacement association with the new table c). If so, execute S7; otherwise, the new table that has established a preliminary replacement association is the new table that has established a final replacement association with the current old table.

[0183] S7, based on the set matching priority rules, perform a replacement judgment on the new table that has established a preliminary replacement association, and determine the new table that has established a final replacement association with the current old table based on the replacement judgment result.

[0184] The specific steps for performing replacement judgment on the newly established table with preliminary replacement associations based on the set matching priority rules are as follows:

[0185] S71, let the current old table (old table A) be the first old table, and let the other old table (old table B) corresponding to the new table that has been matched by other old tables be the second old table;

[0186] Obtain the final matching state of the first old table (old table A) and the second old table (old table B) when applying the most similarity matching rule, and obtain the final matching similarity score S of the first old table. replace The final matching similarity score S of the _new and second old tables replace _old.

[0187] The final matching similarity scores of the first old table and the second old table are the similarity scores corresponding to the final matching state they reached when executing the most similarity matching rule. Taking the above example again, since the program in step S5 executes the program with the old table A and the similar new tables a, b, and c according to the set most similarity matching rule until step S53, S53 is the final matching state of the old table A. The weighted similarity score S of the table titles and contents that meet the criteria of S53 is then calculated. title_cell_Ac The final matching similarity score S of the first old table replace _new, i.e., S replace _new=S title_cell_Ac Similarly, the final matching similarity score S of the old table B can be obtained. replace _old.

[0188] S72, determine the priority status code C of the first old table according to the priority status code determination rules. re _new and the priority status code C of the second old table re _old.

[0189] The priority status code determination rule is as follows:

[0190] If the old table achieves a high title similarity score when applying the most similarity matching rule, thus meeting the requirement of establishing a preliminary replacement association, then C re =3;

[0191] If the final matching state of the old table when performing the most similarity matching rule is such that the weighted similarity score of the table title and content meets the requirements for establishing a preliminary replacement association, then C re =2;

[0192] If the final matching state of the old table when performing the most similarity matching rule is such that the table content similarity meets the requirements for establishing a preliminary replacement association, then C re =1;

[0193] If the final matching state of the old table when performing the most similarity matching rule is that the table title similarity meets the requirements for establishing a preliminary replacement association, then C re =0;

[0194] If the old table fails to match when performing the most similarity matching rule, then C re =-1.

[0195] Using the example above, since the old table A, when matched with similar new tables a, b, and c according to the set most similarity matching rule, reaches a final matching state where the weighted score of the table title and content similarity meets the requirements for establishing a preliminary replacement association, the priority status code of the old table A is C. re _new = 2; Similarly, we can obtain the priority status code C of the old table B. re _old.

[0196] S73, determine the matching priority between the first old table and the second old table based on the final matching similarity score and priority status code of the first old table and the second old table;

[0197] Specifically, determine whether the final matching similarity score and priority status code of the first and second old tables meet condition C. re _new>C re _old, or C re _new=C re _old and S replace _new>S replace If the condition _old is met, it means that the matching priority of the first old table is higher than that of the second old table; otherwise, the matching priority of the first old table is lower than that of the second old table.

[0198] If the matching priority of the first old table is higher than that of the second old table, the new table with which the initial replacement association is established is used as the new table with which the final replacement association is established with the first old table. Simultaneously, the replacement association between this new table and the second old table is canceled, and step S4 is repeated for the second old table until it matches a new new table or does not match a new new table. (During this new round of matching, the second old table repeats step S4 to find a new matching table. The resulting title similarity ranking list {S} is...) title}, Table content similarity sorting list {S cell}, and a list sorted by weighted score for table titles and content similarity {S title_cell} is the set of similarity scores for all remaining new tables after removing all new tables that were previously associated with the second old table (and those that failed to replace the original associated old table). In other words, new tables that were previously associated with the second old table but failed to establish an association need to be removed in this round of matching, and their similarity scores are not included in the sorting list {S}. title}、{S cell}、{S title_cell}middle);

[0199] If the matching priority of the first old table is lower than that of the second old table, then the replacement association between the first old table and the newly established table with which the initial replacement association has been established is cancelled, while the replacement association between the second old table and the new table remains unchanged. Simultaneously, the first old table executes step S4 until it matches a new table or does not match a new table. (Similarly, the first old table repeats step S4 to re-find a new table that matches it.) The resulting title similarity ranking list {S... title}, Table content similarity sorting list {S cell}, and a list sorted by weighted score for table titles and content similarity {S title_cell {S} is the set of similarity scores for all remaining new tables after removing all new tables that were previously associated with the first old table (and those that failed to replace the original associated old table). In other words, new tables that were previously associated with the first old table but failed to establish an association are removed in this round of matching, and their similarity scores are not included in the sorting list. title}、{S cell}、{S title_cell}middle);

[0200] If the first or second old table does not match the new table, it is considered a match failure and will not be replaced.

[0201] Using the example above, assuming that the matching priority of old table A is determined to be higher than that of old table B (i.e., old table A is more similar to new table c than old table B), then the final replacement association is established between new table c and old table A, while the initial replacement association established between new table c and old table B is canceled. Then, old table B repeats step 4 to find a new table that matches it. If repeating step 4 does not find a new table that matches old table B, then old table B is considered to have failed to match and is not replaced.

[0202] Conversely, assuming that the matching priority of old table B is determined to be higher than that of old table A (i.e., old table B is more similar to new table c than old table A), then the initial replacement association established between new table c and old table A is cancelled, while the replacement association between old table B and new table c remains unchanged. Then, old table A executes step S4 to find a new matching table. If repeating step 4 does not find a new matching table with old table A, then old table A is considered to have failed to match, and old table A is not replaced.

[0203] S8, repeat steps S4-S7 for the next old table, until all old tables have been matched.

[0204] S9 performs a unified replacement of the content of all new and old tables that have established a replacement association. This means that all new tables with established replacement associations are updated and replaced with their corresponding old tables (i.e., the entire table is replaced, including the table title, table content, and table cell content of the entire new table). Old tables that fail to match and do not have established replacement associations are not processed or replaced.

[0205] The present invention also provides an apparatus for automatically updating financial professional report table data using the method described above, including a file processing module, a document chapter mapping module, a new and old table matching module, and a new and old table content replacement module.

[0206] The file processing module is used to extract table titles, table chapters, and table content from new and old files, and to normalize all extracted table titles, table chapters, and table content.

[0207] The document chapter mapping module is used to obtain a chapter mapping dictionary between new and old files, using chapters as the mapping scale.

[0208] The new and old table matching module is used to calculate the new and old table similarity between a single old table and all new tables belonging to the same major chapter based on the chapter mapping dictionary, match the new table that meets the most similar matching rule based on the calculated new and old table similarity value, establish a preliminary replacement association between the new table that meets the most similar matching rule and the current old table, and determine whether the new table that has established a preliminary replacement association is a new table that has been matched by other old tables. If so, the new table that has established a preliminary replacement association is replaced according to the set matching priority rule.

[0209] The new and old table content replacement module is used to uniformly replace the content of all new and old tables that have implemented the replacement association.

[0210] The present invention also provides a computer device comprising: a processor and a memory, wherein the memory stores at least one program, the at least one program being loaded and executed by the processor to implement the method described above.

[0211] The present invention also provides a computer-readable storage medium storing at least one program, which is loaded and executed by a processor to implement the method described above.

[0212] The present invention also provides a computer program product comprising at least one program segment stored in a computer-readable storage medium; a processor of a communication device reads the at least one program segment from the computer-readable storage medium, and the processor executes the at least one program segment to cause the communication device to perform the method described above.

[0213] It should be understood that the described embodiments are merely some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

Claims

1. A method for automatically updating data in financial professional report forms, characterized in that, Specifically, the following steps are included: S1 extracts the table titles, the chapters to which the tables belong, and the table content from the new and old files, respectively. The new file refers to the template file, and the old file refers to the file to be modified. S2, normalize all extracted table titles, table chapters, and table contents; S3, using chapters as the mapping scale, obtains the chapter mapping dictionary between new and old files; S4. Based on the chapter mapping dictionary, calculate the new-old table similarity between a single old table and all remaining new tables after removing all new tables that were previously associated with the old table but failed to be associated with it, after removing all new tables that were previously associated with the old table but failed to be associated with it, according to the chapter mapping dictionary. The similarity between the old and new tables includes the title similarity S. title Table content similarity S cell And the weighted score S for similarity between table titles and content. title_cell ; S5. Based on the calculated similarity value between the new and old tables, filter and match the new table that meets the conditions according to the set most similarity matching rule. If no new table that meets the most similarity matching rule is found, the current old table is considered to have failed to match and will not be replaced. Repeat step S4 to match the next old table. If a new table that meets the most similarity matching rule is matched, proceed to step S6. S6, establish a preliminary replacement association between the new table that satisfies the most similar matching rule and the current old table, and then determine whether the new table that has established a preliminary replacement association is a new table that has established a final replacement association with other old tables. If so, execute S7; otherwise, the new table that has established a preliminary replacement association is the new table that has established a final replacement association with the current old table. S7, perform a replacement judgment on the new table that has established a preliminary replacement association according to the set matching priority rules, and determine the new table that establishes a final replacement association with the current old table based on the replacement judgment result; S8. Repeat steps S4-S7 to match the next old table until all old tables have been matched. S9 replaces the content of all new and old tables that have implemented the replacement association.

2. The method for automatically updating financial professional report table data according to claim 1, characterized in that, The specific steps in step S4 for calculating the similarity between a single old table and all remaining new tables after removing all new tables that were previously associated with the old table but failed to be associated with it are as follows: S41, obtain the title subject keywords of the current old table, and match the new tables with the same title subject keywords from all new tables belonging to the same major chapter mapping pair based on the title subject keywords of the current old table. Old tables with old titles that do not contain any subject keywords can only be matched with new tables with new titles that do not contain any subject keywords. The matched new tables are called similar new tables. S42, calculate the title similarity S between the current old table and each similar new table. title Similarity S to table content cell ; S43, Title similarity S between the current old table and each similar new table title Similarity S to table content cell Calculate the weighted score S for the similarity of the corresponding table titles and content. title_cell ; S44, calculate all the title similarity values ​​S title Table content similarity value S cell The weighted score S for similarity between table title and content title_cell Sort the titles according to their scores from highest to lowest to obtain a list of titles with similarity scores {S}. title }, Table content similarity sorting list {S cell } and a list sorted by weighted score of table title and content similarity {S title_cell } 3. The method for automatically updating financial professional report table data according to claim 2, characterized in that, In step S42, the title similarity S between the current old table and a single similar new table is calculated. title The specific steps are as follows: By referring to the set subject keyword list, extract subject keywords from the current old table and similar new tables respectively; The title matching score S is determined based on the similarity between the subject keywords in the current old table and the subject keywords in similar new tables. key ; Obtain the semantic vectors of the table titles of the current old table and similar new tables respectively; The cosine similarity S between the old and new title vectors is calculated based on the semantic vectors of the old and new title vectors. vec Cosine similarity S between old and new title vectors vec As a reference value for title similarity; Based on the title matching score S key Cosine similarity S between the old and new title vectors vec Calculate the title similarity S between the current old table and similar new tables. title .

4. The method for automatically updating financial professional report table data according to claim 3, characterized in that, If the subject keywords in the current old table are exactly the same as the subject keywords in a similar new table, then the title matching score is S. key =1; If the subject keywords in the current old table match the subject keywords in a similar new table, the title matching score is S. key =0.5; If the subject keywords in the current old table do not match the subject keywords in a similar new table, the title matching score will be S. key =0.

5. The method for automatically updating financial professional report table data according to claim 3, characterized in that, The cosine similarity S between the old and new title vectors vec The calculation formula is: Among them, v old v is the semantic vector for the title of the current old table. new For the title semantic vector of a similar new table; The title similarity S between the current old table and similar new tables title The calculation formula is: Among them, w k and w t All of these are set weight values.

6. The method for automatically updating financial professional report form data according to claim 2 or 3, characterized in that, If the current old table has no header, then the header similarity S between the current old table and similar new tables will be calculated. title Set to 0.

7. The method for automatically updating financial professional report table data according to claim 2, characterized in that, In step S42, the table content similarity S between the current old table and a single similar new table is calculated. cell The specific steps are as follows: Calculate the initial similarity S between the contents of the old table (after normalization) and the new table (after normalization). cell_0 ; Where, {O1,…,O n } represents the set of all cell contents of the current old table after normalization, {N1,…,N n } represents the collection of all cell contents after normalization of the similar new table, and LENGTH() returns the length of the collection; Based on the relationship between the smallest section above the current old table and the smallest section above the similar new table, the initial similarity S of the table content is calculated. cell_0 Add the minimum chapter relationship weight to the table to obtain the table content similarity S between the current old table and similar new tables. cell .

8. The method for automatically updating financial professional report table data according to claim 7, characterized in that, If the minimum chapter text of both the current old table and the similar new table is not empty and the minimum chapter text of both tables is contained within each other, then S cell_0 ≥T contain_chapter Then S cell =S cell_0 +W contain_chapter ; If the minimum chapter text of both the current old table and the similar new table is not empty and the minimum chapter text of the two tables is equal, then S cell_0 =0, then S cell =W contain_chapter ; If the minimum chapter text of both the current old table and the similar new table is not empty and the minimum chapter text of the two tables is equal, then S cell_0 If S ≠ 0, then S cell =S cell_0 +W same_chapter ; Among them, T contain_chapter To set the table content similarity threshold, W contain_chapter and W same_chapter These are all set minimum chapter relationship weight values ​​for the tables.

9. The method for automatically updating financial professional report form data according to claim 7 or 8, characterized in that, If the difference between the smallest chapter text of the current old table and the similar new table is a specified keyword, S cell =-1.

10. The method for automatically updating financial professional report table data according to claim 2, characterized in that, In step S5, the specific steps for filtering the new table that meets the conditions based on the calculated similarity value between the old and new tables and according to the set most similarity matching rule are as follows: S51, Determine the title similarity ranking list {S title S, the highest score in} title_max Does it satisfy S? title_max >T title_top If the condition is met, the new table corresponding to the title similarity value will be replaced by the current old table; otherwise, step S52 will be executed. S52, Determine the title similarity ranking list {S title S, the highest score in} title_max Does it satisfy 0? title_max <T title_min If the condition is met, the current old table is considered to have failed to match and will not be replaced. Step S4 is repeated to match the next old table. Otherwise, step S53 is executed.​ S53, Determine the weighted score ranking list of table titles and content similarity {S title_cell S, the highest score in} title_cell_max Does it satisfy S? title_cell_max >T title_cell If the conditions are met, a replacement association is established between the new similar table corresponding to the weighted score of the similarity between the table title and content and the current old table; otherwise, step S54 is executed. S54, Sort the table content similarity list {S cell S, the highest score in} cell_max Does it satisfy S? cell_max <T cell_min If the condition is met, the current old table is considered to have failed to match and will not be replaced. Step S4 is repeated to match the next old table; otherwise, step S55 is executed. S55, Sort the table content similarity list {S cell S, the highest score in} cell_max Does it satisfy S? cell_max >T cell If the condition is met, a replacement association is established between the new similar table corresponding to the table content similarity value and the current old table; otherwise, step S56 is executed. S56, Determine the title similarity ranking list {S title S, the highest score in} title_max Does it satisfy S? title_max >T title If the conditions are met, the new table corresponding to the title similarity value will be linked to the current old table for replacement; otherwise, the current old table will be considered to have failed to match and will not be replaced. This step S4 will be repeated to match the next old table. Among them, T title_top T title_min and T title All are set title similarity thresholds, T title_cell A weighted threshold for the similarity of table titles and content, T cell_min and T cell These are all set table content similarity thresholds.

11. The method for automatically updating financial professional report table data according to claim 10, characterized in that, The T title_min <T title <T title_top T cell_min <T cell .

12. The method for automatically updating financial professional report table data according to claim 10, characterized in that, The title similarity threshold T title_min =0.5, T title =0.829, T title_top =0.98, Table content similarity threshold T cell_min =0.04, T cell =0.082, weighted threshold T for table title and content similarity title_cell =0.

7.

13. The method for automatically updating financial professional report table data according to claim 1, characterized in that, In step S7, the replacement judgment is performed on the new table with which the initial replacement association is established according to the set matching priority rules. The specific steps for determining the new table with which the final replacement association is established with the current old table based on the replacement judgment result are as follows: S71, let the current old table be the first old table, and let the other old tables corresponding to the new table that has been matched by other old tables be the second old table; Obtain the final matching state of the first and second old tables when applying the most similarity matching rule, and obtain the final matching similarity score S of the first old table. replace_ The final matching similarity score S of the new and second old tables replace_ old; S72, determine the priority status code C of the first old table according to the priority status code determination rules. re_ The priority status code for new and second-old tables is C. re_ old; S73, determine the matching priority of the first old table and the second old table based on the final matching similarity score and priority status code of the first old table and the second old table. If the matching priority of the first old table is higher than that of the second old table, take the new table that has established a preliminary replacement association as the new table that has established a final replacement association with the first old table, and cancel the replacement association between the new table and the second old table and repeat step S5 until the second old table matches the new table or does not match the new table. If the matching priority of the first old table is lower than that of the second old table, then cancel the replacement association between the first old table and the new table with which the initial replacement association has been established, keep the replacement association between the second old table and the new table unchanged, and execute step S5 until the first old table matches the new table or does not match the new table. If the first or second old table does not match the new table, the first or second old table that failed to match is considered a failed match and will not be replaced.

14. The method for automatically updating financial professional report form data according to claim 13, characterized in that, The priority status code determination rule is as follows: If the old table achieves a high title similarity score when applying the most similarity matching rule, thus satisfying the requirement to establish a replacement association, then C re =3; If the final matching state of the old table when performing the most similarity matching rule is such that the weighted similarity score of the table title and content meets the requirements for establishing a replacement association, then C re =2; If the final matching state of the old table when performing the most similarity matching rule is such that the table content similarity meets the requirements for establishing a replacement association, then C re =1; If the final matching state of the old table when performing the most similarity matching rule is that the table title similarity meets the requirements for establishing a replacement association, then C re =0; If the old table fails to match when performing the most similarity matching rule, then C re =-1.

15. The method for automatically updating financial professional report form data according to claim 13, characterized in that, The specific steps in step S73 for determining the matching priority between the first old table and the second old table based on the final matching similarity score and priority status code are as follows: Determine whether the final matching similarity score and priority status code of the first and second old tables meet condition C. re_ new>C re_ old, or C re_ new = C re_ old and S replace_ new>S replace_ If the condition is met, it means that the matching priority of the first old table is higher than that of the second old table; otherwise, the matching priority of the first old table is lower than that of the second old table.

16. The method for automatically updating financial professional report table data according to claim 13, characterized in that, The final matching similarity scores of the first old table and the second old table are the similarity scores corresponding to the final matching state they reached when performing the most similar matching rule.

17. An apparatus for automatically updating financial professional report form data using the method described in any one of claims 1-16, characterized in that, include: The file processing module is used to extract table titles, the chapters to which tables belong, and table content from new and old files, and to normalize all extracted table titles, the chapters to which tables belong, and table content. The document chapter mapping module is used to obtain a chapter mapping dictionary between new and old files, using chapters as the mapping scale; The new and old table matching module is used to calculate the new and old table similarity between a single old table and all new tables belonging to the same major chapter mapping pair based on the chapter mapping dictionary. Based on the calculated new and old table similarity values, it matches the new table that meets the most similar matching rule. After establishing a preliminary replacement association between the new table that meets the most similar matching rule and the current old table, it determines whether the new table that has established a preliminary replacement association is a new table that has been matched by other old tables. If so, it performs a replacement judgment on the new table that has established a preliminary replacement association according to the set matching priority rule. The module for replacing content in old and new tables is used to uniformly replace the content in all old and new tables that have been linked for replacement.

18. A computer device, characterized in that, The computer device includes a processor and a memory, the memory storing at least one program, the at least one program being loaded and executed by the processor to implement the method as described in any one of claims 1 to 16.

19. A computer-readable storage medium, characterized in that, The readable storage medium stores at least one program segment, which is loaded and executed by a processor to implement the method as described in any one of claims 1 to 16.

20. A computer program product, characterized in that, The computer program product includes at least one program segment stored in a computer-readable storage medium; a processor of the communication device reads the at least one program segment from the computer-readable storage medium, and the processor executes the at least one program segment, causing the communication device to perform the method as described in any one of claims 1 to 16.

Images