Key Information Extraction Method Based on Business Review

Abstract

An embodiment of the present invention discloses a key information extraction method based on business audit. First, the system collects all key fields in the business audit process, and then establishes a database including all key fields based on all key fields. According to the requirements of different business process types, the key fields that match the process are screened out from the database, and the business process key fields are generated. Finally, the system configures the key fields of users corresponding to the same business process type based on each business process type and the business process key fields corresponding to the business process type. Through database management and process adaptation, key fields are automatically screened and arranged, unnecessary screening work is reduced, and the audit speed is improved. The key information extraction method based on business audit provides automated key field screening and optimizes the arrangement order, so that auditors can focus more on important information and reduce the risk of errors caused by long audits.

Description

Technical Field

[0001] The present invention relates to the technical field of information extraction, and in particular to a key information extraction method based on business audit. Background Art

[0002] In financial industries such as factoring, when handling business, it is usually necessary to conduct layer-by-layer review, including review of contracts, accounts receivable, loan issuance, repayment, customer credit and other aspects, making the review process very cumbersome.

[0003] Each process requires multiple reviewers to approve the fields that need to be checked in the process, and then proceed to the next process for approval. The keywords approved in each process are different, so in actual operation, a business transaction requires multiple review steps to complete, and each review step takes a very long time. All fields in the business are screened and approved one by one, resulting in very low work efficiency.

[0004] In the early stages of the review process, reviewers will use various functions to filter, highlight, and annotate the content that needs to be reviewed to highlight key information that needs to be checked in order to speed up the review process. However, this method of review is still very lengthy and inefficient.

[0005] Another existing approach is to filter out irrelevant fields on the review page before conducting the review. However, different reviewers within the same process have different review styles, making it difficult to standardize and unify the filtering process. This requires adding individual permissions through the back-end interface to accommodate the review styles of different reviewers. This approach requires a significant workload and consumes significant server resources, hindering the processing flow and ultimately preventing improvements in work efficiency.

[0006] Secondly, during the audit process, some fields similar to key information may appear in the audit content. These fields are often filtered out during the filtering process. However, similar fields may lead to errors in the audit results, which may lead to subsequent risks. For example, the difference between "deposit" and "down payment" in a contract may be just one word, representing completely different meanings. During the audit process, each auditor needs to review multiple different fields of key information. Long working hours will affect the auditor's mental state. Eye fatigue will further increase the risk of such errors, resulting in erroneous audit results. Summary of the Invention

[0007] Based on this, it is necessary to propose a key information extraction method based on business audit to address the above problems.

[0008] In a first aspect, a method for extracting key information based on business audit is proposed, the method comprising:

[0009] 101. Collect all key fields in the business review process and establish a database based on all the key fields;

[0010] 102. Configure corresponding key fields for each business process type from the database, and generate business process key fields for each business process type;

[0011] 103. Based on each business process type, configure business process key fields for users of each business process type.

[0012] In at least one embodiment of the present application, the step 101 includes:

[0013] 1011. Obtain historically used key fields in each business review process, where the historically used key fields include: user query key fields, marked key fields, and input key fields;

[0014] 1012. Prioritize the historically used key fields to obtain the prioritized key fields;

[0015] 1013. Establish a database based on the sorted key fields.

[0016] In at least one embodiment of the present application, the step of configuring corresponding key fields for each business process type from the database and generating business process key fields for each business process type includes:

[0017] 1021. Based on the mapping relationship between each business process type and the key fields in the database, configure the corresponding key fields for each business process type;

[0018] 1022. Configure a corresponding key field set according to each business process type to generate business process key fields for each business process type.

[0019] In at least one embodiment of the present application, the steps after step 103 further include:

[0020] 1031. Obtain historical usage records of key fields of the target user account, and reorder the business process key fields according to the historical usage records of key fields to obtain user-specific business process key fields.

[0021] 1032. Configure key fields for the target user account based on the user-specific business process key fields.

[0022] In at least one embodiment of the present application, the step 1031 includes:

[0023] The target user account's key field usage history records include: the target user account's key field usage frequency and key field query frequency;

[0024] 1031a. Reorder the business process key fields according to the usage frequency of the key fields and the query frequency of the key fields to filter out key fields whose usage frequency and / or query frequency exceeds a threshold, and obtain modified key fields;

[0025] 1031b. Configure a priority for the changed key field based on the usage frequency and query frequency of the changed key field. If the usage frequency exceeds the usage threshold, configure a first-level priority. If the query frequency exceeds the query threshold, configure a second-level priority. If the query frequency exceeds the query threshold and the usage frequency exceeds the usage threshold, configure a third-level priority.

[0026] In at least one embodiment of the present application, the steps after step 1032 further include:

[0027] 201. Obtaining the user's access records during the business review process, wherein the access records include: the user's hovering time on a specific non-key field and the number of times the user accesses the specific non-key field;

[0028] 202. Acquire specific non-key fields based on the hovering duration and the number of visits, and generate suspected key fields;

[0029] 203. Generate a suspected priority structure based on the suspected key fields and a database;

[0030] 204. Determine whether the suspected key field needs to be added to the database. If so, add the suspected key field to the database, and configure a priority for the suspected key field added to the database according to the suspected priority structure.

[0031] In at least one embodiment of the present application, the specific steps of step 203 include:

[0032] 2031. Based on the suspected key field and the database, determine whether there is a key field similar to the suspected key field in the key fields of the database; if so, obtain the priority of the similar key field from the database to generate a similar priority;

[0033] 2032. Generate a suspected priority structure based on the original priority structure and the similar priority, and configure priorities for suspected key fields based on the suspected priority structure. The suspected priority structure is the priority structure remaining after removing similar priorities from the original priority structure, and the original priority structure is the priority structure of the key fields in the database.

[0034] In at least one embodiment of the present application, the specific steps of 204 include:

[0035] 2041. Determine whether the hovering duration is within a hovering threshold range. If so, add the suspected key field to the database.

[0036] 2042. If it is not within the hovering threshold range, determine whether the number of visits is within the access threshold; if so, add the suspected key field into the database.

[0037] In at least one embodiment of the present application, the step 204 further includes:

[0038] 301. Obtain the total data of the key business fields reviewed by the target user account during this cycle and generate first usage data;

[0039] 302. Calculate the usage overlap rate between the first usage data and the user-specific business process key fields;

[0040] 303. If the overlap rate is greater than the overlap rate threshold, the ratio of the key fields in each subsequent business to be reviewed is calculated to generate overlap ratio data;

[0041] 304. Filter out the minimum value in the overlap ratio data, take the pending business corresponding to the minimum value as a priority business, and take the priority business as the first business to be processed in the next cycle time.

[0042] In at least one embodiment of the present application, the specific steps of 301 include:

[0043] 3011. Calculate the key fields of each business in the target user account during the current period, generate the number of types of key fields in each business, and obtain first usage data.

[0044] 3012. Calculate proportion data based on the ratio of the number of types of the first usage data to the key fields of the user-specific business process;

[0045] 3013. Accumulate the proportion data within this cycle and calculate the average value to obtain the usage overlap rate.

[0046] Implementing the key information extraction method based on business audit of this embodiment will have at least the following beneficial effects:

[0047] The above-mentioned method for extracting key information based on business audit first collects all key fields in the business audit process, and then establishes a database including all key fields based on all key fields.

[0048] Based on the requirements of different business process types, key fields that match the process are filtered out from the database and business process key fields are generated.

[0049] Finally, the system configures the key fields of users corresponding to the same business process type based on each business process type and the business process key fields corresponding to the business process type.

[0050] Through database management and process adaptation, key fields can be automatically screened and arranged to reduce unnecessary screening work and increase review speed.

[0051] The key information extraction method based on business audit provides automated key field screening and optimizes the arrangement order, allowing auditors to focus more on important information and reduce the risk of errors caused by long audit time.

[0052] It solves the problems of inefficiency and standardization in the traditional audit model, greatly improves audit efficiency, reduces human errors, and enhances the accuracy and adaptability of audits. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0054] in:

[0055] Figure 1 This is a flow chart of a first embodiment of a method for extracting key information based on business audit;

[0056] Figure 2 This is a flow chart of Example 2 of a method for extracting key information based on business audit;

[0057] Figure 3 This is a flow chart of Example 3 of a method for extracting key information based on business audit;

[0058] Figure 4 This is a flow chart of Example 4 of a method for extracting key information based on business audit;

[0059] Figure 5 A flowchart of another embodiment of a method for extracting key information based on business audit;

[0060] Figure 6 This is a flow chart of Example 5 of the key information extraction method based on business audit. DETAILED DESCRIPTION

[0061] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0062] This embodiment provides a method for extracting key information based on business audit, the method comprising:

[0063] 101. Collect all key fields in the business review process and establish a database based on all the key fields;

[0064] 102. Configure corresponding key fields for each business process type from the database, and generate business process key fields for each business process type;

[0065] 103. Based on each business process type, configure business process key fields for users of each business process type.

[0066] Example 1:

[0067] Please refer to Figures 1-6 In this embodiment, the system first collects all key fields in the business review process, and then establishes a database containing all key fields based on all key fields.

[0068] Based on the requirements of different business process types, the key fields that match the process are filtered out from the database, and the business process key fields corresponding to each business process are generated.

[0069] Finally, the system configures the key fields of users corresponding to the same business process type based on each business process type and the business process key fields corresponding to the business process type.

[0070] Through database management and business process adaptation, key fields can be automatically screened and arranged to reduce unnecessary screening work and increase review speed.

[0071] The key information extraction method based on business audit provides automated key field screening and optimizes the arrangement order, allowing auditors to focus more on important information and reduce the risk of errors caused by long audit time.

[0072] It solves the problems of inefficiency and standardization in the traditional audit model, greatly improves audit efficiency, reduces human errors, and enhances the accuracy and adaptability of audits.

[0073] In at least one embodiment of the present application, the step of collecting all key fields in the business review process and establishing a database based on all key fields includes:

[0074] 1011. Obtain historically used key fields in each business review process, where the historically used key fields include: user query key fields, marked key fields, and input key fields;

[0075] 1012. Prioritize the historically used key fields to obtain the prioritized key fields;

[0076] 1013. Establish a database based on the sorted key fields.

[0077] Example 2:

[0078] Please refer to Figure 2 In this embodiment, during the review process, the system first collects the key fields that the user has queried, marked, and input in the historical usage records to form a historical usage key field set.

[0079] The system obtains the historically used key fields in each business review process, and then prioritizes the historically used key fields. It counts the usage frequency, query frequency, and number of manual interventions (manual interventions include manual modification, manual marking, etc.) of the key fields, and performs weighted sorting on the key fields based on these parameters to ensure that important fields have higher priority.

[0080] After prioritizing the historically used key fields, the sorted key fields are obtained. The system establishes a database based on the sorted key fields to ensure that subsequent review processes can be quickly called and matched.

[0081] By automatically recording user interaction behaviors and automatically extracting historical key fields, the accuracy of data collection can be improved.

[0082] The sorting of key fields is dynamically adjusted based on usage frequency and query frequency to improve audit efficiency and accuracy.

[0083] By storing and dynamically adjusting key fields in the database, the system can adapt to different business scenarios and improve the intelligence level of the system.

[0084] Key fields are phrases, words or sentences that exist in the audit process; business process key fields are key field tables or data that need to be referenced by each business process.

[0085] In at least one embodiment of the present application, the step of configuring corresponding key fields for each business process type from the database and generating business process key fields for each business process type includes:

[0086] 1021. Based on the mapping relationship between each business process type and the key fields in the database, generate the key fields corresponding to each business process type configuration;

[0087] 1022. Generate business process key fields according to the key field set corresponding to each business process type configuration.

[0088] Please refer to Figure 2 In this embodiment, the system first establishes a mapping relationship with the database according to different business processes, so that each business process has a mapping relationship with the database, and obtains the historical usage key fields of different business processes through the mapping relationship, and classifies the obtained historical key fields according to the business process type, and summarizes the classified historical key fields, so that different business processes can automatically filter out the corresponding key fields from the database to provide the storage location of the historical usage key fields used by each business process.

[0089] The system then filters out the corresponding key fields in the database based on each business process type.

[0090] Generate business process key fields based on the key fields corresponding to each business process type configuration.

[0091] Business process key fields are a set of key fields corresponding to each business process type.

[0092] Through the key fields of the business process, the key fields of this business process stage are intelligently highlighted, filtered and sorted to improve the review efficiency.

[0093] Automatically filter out key fields through database matching, reduce manual screening work, and improve audit speed. Through intelligent screening and highlighting of key fields, it can reduce audit errors caused by visual fatigue.

[0094] By using the historical key fields of the business process, the business process key fields are uniformly configured for each user in the business process, thereby providing each user in the business process with a unified business process key field to assist in subsequent audits.

[0095] Historical usage key fields are based on obtaining key fields in all audit records of each business review process, including key fields for user queries, marked key fields, and input key fields. Priority sorting is based on the number of times each key field appears. For example, key fields with a high number of appearances are given a higher priority (fifth-level priority), and key fields with a low number of appearances are given a lower priority (first-level priority).

[0096] It should be noted that in this embodiment, the mapping relationship between each business process type and the key field in the database is a one-to-one correspondence. In other embodiments, a one-to-many, many-to-many, or many-to-one relationship can be adopted; the business process type is the type of review stage in which each business process is located, such as: preliminary review, review, amount review, etc.; the key field set is the set of all key fields in the historically used key fields corresponding to each business process type.

[0097] For example, business process types may include contract review (involving fields such as contract number, amount, and payment terms), accounts receivable review (involving fields such as payment period and payer credit information), and payment collection review (involving fields such as payment amount, payment time, and payment voucher). The system first identifies the current business process, then queries the database to extract all stored key fields and their corresponding priorities.

[0098] Based on the matching rules between the preset business process type and key fields, filter out the key fields applicable to the current business process.

[0099] In the audit system, the core fields of the business process are highlighted and irrelevant fields are hidden to reduce the cognitive burden on auditors. The most important fields are placed prominently in the audit interface based on frequency of use.

[0100] Include high priority fields such as amount, contract number, etc.

[0101] The order of fields may be adjusted based on business logic, such as displaying frequently used fields first.

[0102] Only the fields most relevant to the current business process are displayed, reducing the time auditors spend processing irrelevant information.

[0103] Avoid missing important fields and improve the accuracy of the review.

[0104] Through intelligent sorting, reviewers can quickly focus on the most important review content.

[0105] In at least one embodiment of the present application, the steps after step 103 further include:

[0106] 1031. Obtain historical usage records of key fields of the target user account, and reorder the business process key fields according to the historical usage records of key fields to obtain user-specific business process key fields.

[0107] 1032. Configure key fields for the target user account based on the user-specific business process key fields.

[0108] Example 3:

[0109] Please refer to Figure 2 、 Figure 3 In this embodiment, the system obtains the historical usage records of key fields of the target user account during the review process (wherein the historical usage records of key fields include the frequency of querying a certain field, the frequency of query, etc.), and then re-sorts the business process key fields according to the historical usage records of key fields (for example, based on the frequency of use), and obtains the re-sorted business process key field list for the current user - that is, the user-specific business process key fields, and configures the user-specific business process key fields to the current user.

[0110] The system automatically records the user's interaction with key fields during the review process, thereby forming a unique historical usage record for the user. Then, based on the user's historical usage record, the key fields are reordered to ensure that the reordered business process key fields can better match the current target user.

[0111] Since each user has different audit habits and preferences, user behavior data analysis is used to adjust the sorting priority of key business fields, so that each user can perform audits based on the most suitable audit habits, thereby improving audit efficiency.

[0112] By obtaining the historical usage records of the key fields of the target user account, and then reordering the key fields in the initially configured business process key fields based on the query or usage frequency of the key fields in the historical usage records, the user's usage habits in his or her own usage records are obtained, and then the priority sorting of the key fields in the initially configured business process fields is adjusted based on the target user account's own historical usage records. Based on the preliminary configuration, from the perspective of the target user, user-specific business process key fields suitable for target use can be deployed, so that the user-specific business process key fields can be better matched with the target user, thereby improving the review speed of the target user and making the target user review more efficient and accurate.

[0113] The target user account is the account corresponding to one of the users in the manual or machine selection process type; the historical usage record is the usage frequency and query frequency of key fields in the target user account during the business review process.

[0114] For example:

[0115] The key fields of the target user account are: deposit first-level priority, fine second-level priority, amount first-level priority, and loan first-level priority.

[0116] Historical usage records of key fields in the target user account: Deposit usage frequency is 3 times and query frequency is 6 times, Amount usage frequency is 5 times and query frequency is 1 time, Loan usage frequency is 8 times and query frequency is 7 times.

[0117] The query threshold is 5 times, and the usage threshold is 6 times;

[0118] The key field "Deposit" is configured as the second-level priority, the key field "Amount" is configured as the first-level priority, and the key field "Loan" is configured as the third-level priority.

[0119] The user-specific business process key fields are the second-level priority for deposit, the first-level priority for amount, the third-level priority for loan, and the second-level priority for fine.

[0120] Secondly, as users continue to use the system, their historical usage records will be continuously updated. The system can adjust the order of fields based on new behaviors to continuously adapt to users' changing needs.

[0121] In at least one embodiment of the present application, the step 1031 includes:

[0122] The user's key field usage history records include: the usage frequency of the key field and the query frequency of the key field;

[0123] 1031a. Reorder the business process key fields according to the usage frequency of the key fields and the query frequency of the key fields to filter out key fields whose usage frequency and / or query frequency exceeds a threshold, and obtain modified key fields;

[0124] 1031b. Configure a priority for the changed key field based on the usage frequency and query frequency of the changed key field. If the usage frequency exceeds the usage threshold, configure a first-level priority. If the query frequency exceeds the query threshold, configure a second-level priority. If the query frequency exceeds the query threshold and the usage frequency exceeds the usage threshold, configure a third-level priority.

[0125] Please refer to Figure 3In this embodiment, the system is based on each user's behavior data in the historical usage record of the key fields of the target user account, including the usage frequency and query frequency of each key field.

[0126] Through background logs or user operation traces, statistics are collected and historical usage records of each key field are generated.

[0127] The system reorders business process key fields based on the usage frequency and query frequency of the key fields. When the usage frequency exceeds the usage threshold, the key field is marked as a changed key field; when the query frequency of the key field exceeds the threshold, the key field is marked as a changed key field; a first-level priority is configured for the changed key fields corresponding to the usage frequency exceeding the usage threshold; a second-level priority is configured for the changed key fields that exceed the query threshold; when it is detected that the query frequency exceeds the query threshold and the usage frequency exceeds the usage threshold, a third-level priority is configured for the changed key fields.

[0128] By configuring a higher priority for key fields with high usage frequency and / or query frequency, users are reminded to pay more attention to such key fields, so as to assist users in paying more attention to such key fields during the review process. By extracting the usage frequency and query frequency of the corresponding key fields from the historical usage records of the target users, and then adjusting the priority order of the key fields corresponding to these data based on these data (the usage frequency and query frequency of the key fields), the target users are reminded to pay more careful attention to these key fields that need to be paid attention to during the review process, thereby improving the accuracy of users in the review process.

[0129] Users no longer need to repeatedly search and scroll pages to find these key fields, which significantly improves the efficiency of the review.

[0130] Secondly, by setting a higher priority for changing key fields, the error rate of review is reduced to improve the accuracy of user review.

[0131] It should be noted that the frequency of use of key fields refers to how often users use a specific field during the actual review process. For example, if a field is modified, filled in, or updated in multiple reviews, the frequency of use of this field is high.

[0132] Key field query frequency: This refers to the number of times users query a field during the review process. The query frequency of a field represents the user's level of attention to the field and its importance during the review process.

[0133] In at least one embodiment of the present application, the steps after step 1032 further include:

[0134] 201. Obtaining the user's access records during the business review process, wherein the access records include: the user's hovering time on a specific non-key field and the number of times the user accesses the specific non-key field;

[0135] 202. Acquire specific non-key fields based on the hovering duration and the number of visits, and generate suspected key fields;

[0136] 203. Generate a suspected priority structure based on the suspected key fields and a database;

[0137] 204. Determine whether the suspected key field needs to be added to the database. If so, add the suspected key field to the database, and configure a priority for the suspected key field added to the database according to the suspected priority structure.

[0138] In at least one embodiment of the present application, the specific steps of determining whether the suspected key field needs to be added to the database, and if so, adding the suspected key field to the database, and configuring a priority for the suspected key field added to the database include:

[0139] 2041. Determine whether the hovering duration is within a hovering threshold range;

[0140] 20411. If it is within the hover threshold, the suspected key field is added to the database;

[0141] 2042. If it is not within the hovering threshold range, determine whether the number of visits is within the visit threshold;

[0142] 20421. If so, add the suspected key fields into the database.

[0143] Example 4:

[0144] Please refer to Figure 4 In this embodiment, during the business review process, the system obtains the user's access records during the business review process.

[0145] Analyze the user's hovering time on a specific non-key field and the number of times the user visited the specific non-key field based on the access records.

[0146] The system compares the hovering duration with the hovering threshold range. If the hovering duration is within the hovering threshold range, the specific non-key field is recorded as a suspected key field and added to the database.

[0147] If it is not within the hovering threshold, the system further determines whether the number of visits is within the access threshold. If so, the specific non-key field is recorded as a suspected key field and added to the database.

[0148] By collecting and analyzing user behavior data, the system can automatically identify and update key fields without relying on human intervention, making the system more adaptive.

[0149] By automatically identifying and adding suspected key fields, important information is prioritized, thereby improving efficiency and accuracy during the audit process.

[0150] It should be noted that if the hovering time exceeds the hovering threshold range, it may be that the user has left the workstation for a long time and the mouse pointer has not moved. At this time, the number of visits is further judged. If the number of visits exceeds the access threshold, it means that this is a field that needs special attention. The suspected key field will be added to the database, and the priority of the suspected key field added to the database will be configured to automatically update the key field to avoid the user having to review it separately each time, and the system cannot display or recommend it.

[0151] It should be further explained that if the number of visits does not exceed the visit threshold, it means that the user has not paid attention to the key field here.

[0152] The access threshold, query threshold, and usage threshold are all manually set. The number of visits is the number of times a target user hovers over a key field with their mouse. The query frequency is the number of times a target user queries a key field using the query function. The usage frequency is the number of times a target user enters or marks a key field during the business review process.

[0153] The specific non-key fields are key fields or a sentence or phrase marked by a mouse pointer in the historical usage records of the target user account, such as records generated by circling, moving back and forth, marking, highlighting, etc. with a mouse pointer, and the key fields in the above records are extracted according to the extraction algorithm.

[0154] The hover duration is the time the mouse is hovered. If it exceeds the hover threshold, it is determined that the user corresponding to the target user account has left the workstation or other circumstances.

[0155] The number of visits is then used to determine whether it is within the access threshold, thereby avoiding the inability to add key fields to the database due to system misjudgment, thereby improving the accuracy of the audit process.

[0156] In at least one embodiment of the present application, the specific step of 203 further includes:

[0157] 2031. Based on the suspected key field and the database, determine whether there is a key field similar to the suspected key field in the key fields of the database; if so, obtain the priority of the similar key field from the database to generate a similar priority;

[0158] 2031. Generate a suspected priority structure based on the original priority structure and the similar priority, and configure priorities for suspected key fields based on the suspected priority structure. The suspected priority structure is the priority structure remaining after removing similar priorities from the original priority structure, and the original priority structure is the priority structure of the key fields in the database.

[0159] Embodiment 5:

[0160] Please refer to Figure 6 In this embodiment, the system searches for a key field with high similarity to the suspected key field in the database. If it exists, the system obtains the priority status of the key field with high similarity from the database and generates a similar priority.

[0161] After the system obtains the structure of the priority levels in the database, it obtains the original priority structure, eliminates similar priorities in the priority levels, and obtains a suspected priority structure. In the suspected priority structure, one is selected as the suspected key field for configuration, thereby distinguishing the suspected key field from other key fields in the database to improve the accuracy of the audit.

[0162] At the same time, suspected key fields configured with similar priorities are added to the database to facilitate subsequent audits, thereby improving the accuracy and efficiency of the subsequent audit process.

[0163] Suspected key fields are key fields that target users frequently access or use during the business audit process. By extracting key fields from historical usage records, we can adjust individual users' usage habits of key fields and then adjust individual users' business process key fields to obtain user-exclusive business process key fields, thereby matching individual users' operating habits and further improving the accuracy and efficiency of individual user audits.

[0164] For example:

[0165] If the suspected key field is: deposit;

[0166] The original priority structure includes: first-level priority, second-level priority and third-level priority. The database contains a key field: deposit, and the priority corresponding to the deposit is - second-level priority;

[0167] Then deposit and down payment are key fields with high similarity;

[0168] The suspected priority structure is: first-level priority and third-level priority;

[0169] And configure the first-level priority or the third-level priority for the deposit. In this case, the configured first-level priority or the third-level priority is similar to the priority to distinguish the deposit from the deposit.

[0170] By intelligently and automatically assigning priorities to suspected key fields, it ensures that the sorting and processing order of fields always meets the needs of business processes, avoiding manual intervention and human errors.

[0171] Through similarity analysis and automatic priority configuration, the system can dynamically optimize the audit process so that key fields are given priority in the process, thereby improving overall work efficiency and audit accuracy.

[0172] By identifying similar fields, the system can effectively reduce redundant fields and avoid processing duplicate information, thereby improving the overall performance of the system and reducing interference from irrelevant information.

[0173] The key information extraction method based on business audit provides automated key field screening and optimizes the arrangement order, allowing auditors to focus more on important information and reduce the risk of errors caused by long audit time.

[0174] It solves the problems of inefficiency and standardization in the traditional audit model, greatly improves audit efficiency, reduces human errors, and enhances the accuracy and adaptability of audits.

[0175] In at least one embodiment of the present application, the specific step of 204 further includes:

[0176] 301. Obtain the total data of the key business fields reviewed by the target user account during this cycle and generate first usage data;

[0177] 302. Calculate the usage overlap rate between the first usage data and the user-specific business process key fields;

[0178] 303. If the overlap rate is greater than the overlap rate threshold, calculate the ratio of key fields in each business to be reviewed within the next cycle time to generate overlap ratio data;

[0179] 304. Filter out the minimum value in the overlap ratio data, take the pending business corresponding to the minimum value as a priority business, and take the priority business as the first business to be processed in the next cycle time.

[0180] Please refer to Figure 5 In this embodiment, the system obtains all audited businesses of the target user account within the current cycle time, and then extracts the total data of key fields corresponding to all audited businesses, and generates the first usage data after calculation.

[0181] Then, the usage overlap rate is calculated between the first usage data and the user-specific business process key fields. The usage overlap rate can be used to determine the proportion of all audited businesses relative to the user-specific business process key fields within this cycle.

[0182] When the usage overlap rate is greater than the overlap rate threshold, the overlap ratio of each pending business in the subsequent pending businesses relative to the user-specific business process key fields is calculated to obtain overlap ratio data.

[0183] Then, the minimum value is filtered out from the overlap ratio data, and priority processing business is generated according to the pending business corresponding to the minimum value, and the priority processing business is assigned to the first business to be processed in the next cycle time.

[0184] This will avoid the auditors corresponding to the target user accounts from experiencing audit fatigue due to a high overlap rate, which in turn will lead to a reduction in audit accuracy due to audit fatigue. This will further optimize the overall work efficiency and improve the accuracy of the business audit process.

[0185] For example:

[0186] During this cycle, the number of types of key fields of business No. 1 relative to the user-exclusive business process key fields is 9, the number of types of key fields of business No. 2 relative to the user-exclusive business process key fields is 10, the number of types of key fields of business No. 3 relative to the user-exclusive business process key fields is 8, and the number of types of key fields of business No. 4 relative to the user-exclusive business process key fields is 12. The first usage data is 9, 10, 8, and 12.

[0187] The number of types of user-specific process key fields is 14, and the proportions are: 0.64, 0.71, 0.57, and 0.85, so the usage overlap rate is 0.69.

[0188] The overlap rate threshold is 0.6.

[0189] If 0.69>0.6, the ratio of the key fields in each subsequent business to be reviewed is calculated to generate overlapping ratio data. For example, if the overlapping ratio data is: 0.8, 0.7, 0.6, 0.75, 0.83, 0.4, 0.96, 0.3, 0.87, 0.57, the business to be reviewed corresponding to 0.3 is screened out and the business to be reviewed is arranged as the first business to be processed in the next cycle time.

[0190] In at least one embodiment of the present application, the specific steps of obtaining total data of business key fields reviewed by the target user account within the current cycle, generating first usage data, and calculating the usage overlap rate between the first usage data and the user-specific business process key fields include:

[0191] 3011. Calculate the key fields of each business in the target user account during the current period, generate the number of types of key fields in each business, and obtain first usage data.

[0192] 3012. Calculate proportion data based on the ratio of the number of types of the first usage data to the key fields of the user-specific business process;

[0193] 3013. Accumulate the proportion data within this cycle and calculate the average value to obtain the usage overlap rate.

[0194] Please refer to Figure 5 In this embodiment, the system calculates the number of types of key fields of each reviewed business by the target user account during this cycle, obtains the first usage data based on the number of types of each reviewed business, and provides a data source for subsequent usage overlap rate calculation.

[0195] Then, based on the ratio of the first usage data to the number of types of key fields in the user-specific business process, the overlap rate of each processed business process (within this cycle) is calculated to obtain the proportion data. Then, all the proportion data within this cycle are accumulated and averaged to obtain the usage overlap rate, which is used to measure the ratio between all key field types in the business reviewed by the user within this cycle and the key field types of the user-specific business process to obtain the usage overlap rate. By using the overlap rate, it is further determined whether the reviewer corresponding to the target user is in the fatigue stage, and then subsequent adjustments are made to the business to alleviate the fatigue state, thereby improving the accuracy of the business review.

[0196] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0197] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

Claims

1. A key information extraction method based on business audit, characterized in that: The method comprises: Collect all key fields in the business review process and build a database based on all the key fields; Configure the corresponding key fields for each business process type from the database and generate the business process key fields for each business process type; Based on each business process type, configure business process key fields for each user of that business process type; Obtaining historical usage records of key fields of the target user account, and reordering the business process key fields according to the historical usage records of key fields to obtain user-specific business process key fields; Configure key fields for the target user account based on the user-specific business process key fields; Obtaining the user's access records during the business review process, wherein the access records include: the user's hovering time on a specific non-key field and the number of times the user accesses the specific non-key field; According to the hovering duration and the number of visits, a specific non-key field is obtained, and a suspected key field is generated; generating a suspected priority structure based on the suspected key fields and a database; Determine whether the suspected key field needs to be added to the database, and if so, add the suspected key field to the database, and configure a priority for the suspected key field added to the database according to the suspected priority structure; Obtain the total data of key business fields reviewed by the target user account during this cycle and generate the first usage data; Calculating a usage overlap rate between the first usage data and the user-specific business process key fields; If the overlap rate is greater than the overlap rate threshold, the ratio of key fields in each subsequent business to be reviewed is calculated to generate overlap ratio data; The minimum value in the overlap ratio data is screened out, the pending business corresponding to the minimum value is taken as a priority business, and the priority business is taken as the first business to be processed in the next cycle time.

2. The key information extraction method based on business audit according to claim 1 is characterized in that: The steps of collecting all key fields in the business review process and establishing a database based on all the key fields include: Obtaining historically used key fields in each business review process, wherein the historically used key fields include: user query key fields, marked key fields, and input key fields; Prioritizing the historical usage key fields to obtain sorted key fields; A database is established based on the sorted key fields.

3. The key information extraction method based on business audit according to claim 1 is characterized in that: The steps of configuring corresponding key fields for each business process type from the database and generating business process key fields for each business process type include: According to the mapping relationship between each business process type and the key fields in the database, configure the corresponding key fields for each business process type; A corresponding key field set is configured according to each business process type to generate business process key fields for each business process type.

4. The key information extraction method based on business audit according to claim 1 is characterized in that: The steps of obtaining the target user account's key field usage history records, reordering the business process key fields according to the key field usage history records, and obtaining the user-specific business process key fields include: The target user account's key field usage history records include: the target user account's key field usage frequency and key field query frequency; Reordering the business process key fields according to the usage frequency of the key fields and the query frequency of the key fields to filter out key fields whose usage frequency and / or query frequency exceeds a threshold, and obtaining modified key fields; Based on the usage frequency and query frequency of the modified key field, a priority is configured for the modified key field. If the usage frequency exceeds the usage threshold, a first-level priority is configured. If the query frequency exceeds the query threshold, a second-level priority is configured. If the query frequency exceeds the query threshold and the usage frequency exceeds the usage threshold, a third-level priority is configured.

5. The key information extraction method based on business audit according to claim 1 is characterized in that: The specific steps of generating a suspected priority structure based on the suspected key fields and the database include: Based on the suspected key field and the database, determine whether there is a key field similar to the suspected key field in the key fields of the database; if so, obtain the priority of the similar key field from the database to generate a similar priority; Based on the original priority structure and the similar priority, a suspected priority structure is generated, and priorities are configured for suspected key fields based on the suspected priority structure. The suspected priority structure is the priority structure remaining after removing similar priorities from the original priority structure, and the original priority structure is the priority structure of the key fields in the database.

6. The key information extraction method based on business audit according to claim 1 is characterized in that: The specific steps of determining whether the suspected key field needs to be added to the database, and if so, adding the suspected key field to the database, and configuring the priority of the suspected key field added to the database according to the suspected priority structure include: Determine whether the hovering duration is within a hovering threshold range, and if so, add the suspected key field into the database; If it is not within the hovering threshold range, it is determined whether the number of visits is within the access threshold. If so, the suspected key field is added to the database.

7. The key information extraction method based on business audit according to claim 1 is characterized in that: The specific steps for obtaining the total data of key business fields reviewed by the target user account during this cycle and generating the first usage data include: Calculate the key fields of each business in the target user account during the current period, generate the number of types of key fields in each business, and obtain first usage data; Calculate proportion data based on the ratio of the number of types of the first usage data to the key fields of the user-specific business process; The usage overlap rate is obtained by accumulating and averaging the proportion data within the current cycle.

Images