AI-based system operation intelligent assistant method and system

Abstract

The present invention belongs to the field of system operation technology, and specifically relates to an AI-based system operation intelligent assistant method and system. It includes the following steps: S1. After establishing a new system, establish an operation process; sort out the corresponding initial database of solutions based on the error points in the initial error point database; S2. Determine whether the data information filled in by the user is correct; S3. Prioritize the display of error points and solutions that meet the preset matching degree; S4. Repeat step S2 after entering the next step; S5. Provide multiple matching results for the user to judge; S6. Based on the fuzzy matching results that solve the problem, re-optimize the fuzzy matching of the data information filled in by the user so that the user's choice reaches the preset matching degree and updates the fuzzy matching solution; S7. Match in the starting point database according to the results of the questionnaire; S8. Display the second matching result to the user. If the problem is solved, execute step S6; if not, execute step S9; S9. Connect to manual guidance. This method has a high degree of intelligence.

Description

Technical Field

[0001] The present invention relates to the field of system operation technology, and in particular to an AI-based system operation intelligent assistant method and system. Background Art

[0002] The description of the background technology in the present invention belongs to the related technology related to the present invention and is only used to illustrate and facilitate the understanding of the invention content of the present invention. It should not be understood that the applicant explicitly believes or infers that the applicant believes that it is the prior art of the present invention on the filing date of the first application.

[0003] After the company's new business system goes online, various issues related to the business system, such as system operation training, daily operation guidance, and solving abnormal interactions with peripheral systems, are very resource-consuming and time-consuming. Therefore, based on this demand background, we can introduce a system intelligent assistant to meet the above needs, helping the company to solve the above problems quickly, effectively and at a low cost. Summary of the Invention

[0004] The purpose of the embodiments of the present invention is to provide an AI-based system operation intelligent assistant method and system. The method of the present invention adopts an artificial intelligence learning method to continuously update the database and fuzzy matching scheme to make the guidance closer to the user's needs, so that the system is continuously improved, and the guidance on user's incorrect use is more accurate, saving time and cost.

[0005] The purpose of the embodiment of the present invention is achieved through the following technical solutions:

[0006] An AI-based system operation intelligent assistant method includes the following steps:

[0007] S1. After establishing a new system, establish operational procedures. Based on the operational procedures, predict problems for the new system and conduct system usage investigation and testing. Summarize the predicted and investigated error points to create an initial error database. Based on the error points in the initial error database, create an initial database of corresponding solutions.

[0008] S2 determines whether the data filled in by the user is correct. If correct, proceed to the next step. If wrong, the data filled in by the user is fuzzy matched with the initial database of error points; if there is a fuzzy match result, execute step S3; if no match result is found, execute step S7;

[0009] S3. Prioritize the error points and solutions that meet the preset matching degree; if the user confirms that the problem is solved and correctly proceeds to the next step, execute step S4; if the user's problem is not solved, execute step S5;

[0010] S4. Repeat step S2 after entering the next step;

[0011] S5. Display multiple matching results in order of matching degree and highlight the matching points of the initial error point database and the data filled in by the user for the user to judge; if the user problem is solved, execute step S6; if the user problem is not solved, execute step S7;

[0012] S6. Based on the fuzzy matching results of the problem, the fuzzy matching optimization is performed again on the data entered by the user so that the user's selection reaches the preset matching degree and the fuzzy matching solution is updated;

[0013] S7. Display a questionnaire to the user regarding the incorrect data provided by the user. After the user completes the questionnaire, the user performs a match against the starting point database based on the questionnaire results. If a second matching result is found that is different from the previous matching result, the user proceeds to step S8; if no matching result is found, the user proceeds to step S9.

[0014] S8. Display the second matching result to the user. If the problem is solved, execute step S6; if not, execute step S9;

[0015] S9. Connect to manual guidance, complete the input of data according to manual guidance and successfully enter the next step, establish a new matching result according to the manual solution process, update the initial error point database and the corresponding solution database and execute step S6.

[0016] Furthermore, step S41 is further included after step S4. According to the fuzzy matching result of solving the problem, the data information filled in by the user is fuzzy matched and optimized again so that the user's choice becomes the solution with the highest matching degree, and the fuzzy matching solution is updated.

[0017] Furthermore, in step S5, if the user problem is solved, step S6 is executed; before executing step S6, step S51 is also included. The user is asked whether the matching points between the highlighted initial error point database and the data information filled in by the user are correct. If not, the matching points between the initial error point database and the data information filled in by the user pointed out are compared and analyzed with the originally highlighted matching points, and the matching relationship is re-established based on the matching points pointed out by the user, so that the matching points pointed out by the user are displayed first.

[0018] Furthermore, step S6 also includes step S61 after updating the fuzzy matching scheme, performing a comparison analysis based on the initial error point database pointed out by the user and the matching points of the data filled out by the user and the originally highlighted matching points, re-analyzing and performing fuzzy matching optimization according to the matching points pointed out by the user so that the solution selected by the user becomes the solution with the highest matching degree and the fuzzy matching scheme is updated.

[0019] Furthermore, if there is a second matching result different from the previous matching result in step S7, before executing step S8, the step of applying the problem-solving method of the questionnaire to each step of the system operation is also included, establishing new problem points based on the questionnaire and updating the initial error point database.

[0020] Furthermore, in steps S6 to S9, the fuzzy matching scheme, the initial database of error points, and the corresponding solution database are updated in a label manner, and the steps are as follows:

[0021] (1) storing each error point, fuzzy matching scheme, and corresponding solution in the error point initial database into a tab page, wherein the tab page includes a title, a body, and a tab bar;

[0022] (2) The main body stores error points, fuzzy matching solutions, and corresponding solutions, and stores updated error points, fuzzy matching solutions, and corresponding solutions in the form of revision annotations in the tag column, wherein the revision annotations are in the form of hyperlinks;

[0023] (3) Clicking on the hyperlink of the revision annotation generates a new tab page, which contains the new error points, fuzzy matching schemes, and corresponding solutions after the revision annotation;

[0024] (4) The generated new error points, fuzzy matching schemes, and corresponding solutions are stored in the main body of a new tab page, with the title set to a subpage of the original tab page title. The subpages are marked with features for distinguishing each subpage, including: time, reason for revision, and number of users targeted by the revision;

[0025] (5) Repeat step (2) on each subpage.

[0026] Furthermore, after the user completes the use of the new system, a statistical analysis is performed on the data filled in by the user to find defects in the data filled in by the user in the specific business. If the defective item does not reach the set standard for affecting normal work, the defective item will continue to be recorded; if the defective item has reached the warning value for affecting normal work, the system operation process will be re-specified based on the defective item analysis and the defective item solution to obtain a new process, and steps S1-S9 will be re-executed according to the new process.

[0027] Furthermore, the monitoring system checks whether there are any matters that require approval. If so, the early warning channel reminds the user to complete the approval process of the business document.

[0028] Furthermore, the warning channels include SMS, enterprise WeChat and pop-up services.

[0029] An AI-based system operation intelligent assistant system is used to implement the above assistant method, including:

[0030] Operating system, including operating procedures and operation tutorials;

[0031] Initial database: includes an initial database of error points and a corresponding solution database, used to store initial error points and corresponding solutions sorted out based on the error points in the initial error point database;

[0032] The judgment module is used to judge whether the user operation process is correct;

[0033] Guidance module: if the user operates incorrectly, it will analyze the error and provide solutions for the user to choose;

[0034] The updating module is configured to update the initial error point database and the corresponding solution database according to the user's feedback in the guidance to form a new database.

[0035] The embodiments of the present invention have the following beneficial effects:

[0036] The method of the present invention adopts an artificial intelligence learning method to continuously update the database and fuzzy matching scheme to make the guidance closer to the user's needs, so that the system is continuously improved, and the guidance on user errors is more accurate, saving time and cost. DETAILED DESCRIPTION

[0037] The present application will be further described below with reference to the embodiments.

[0038] To more clearly illustrate the embodiments of the present invention or technical solutions in the prior art, different "one embodiment" or "embodiment" in the following description do not necessarily refer to the same embodiment. Different embodiments may be replaced or combined. Those skilled in the art can also derive other implementation methods based on these embodiments without inventive effort.

[0039] An AI-based system operation intelligent assistant method includes the following steps:

[0040] S1. After establishing a new system, establish operational procedures. Based on the operational procedures, predict problems for the new system and conduct system usage investigation and testing. Summarize the predicted and investigated error points to create an initial error database. Based on the error points in the initial error database, create an initial database of corresponding solutions.

[0041] There are two initial ways to generate error points here. One is based on intelligent prediction, and the other is investigation and testing before the system is officially launched. The two methods are combined to obtain a more comprehensive picture of possible error points. Corresponding solutions are given for the above possible error points, which is convenient for dealing with problems that may arise in subsequent system use.

[0042] S2 determines whether the data filled in by the user is correct. If correct, proceed to the next step. If wrong, the data filled in by the user is fuzzy matched with the initial database of error points; if there is a fuzzy match result, execute step S3; if no match result is found, execute step S7;

[0043] Based on the information filled in by the user, determine whether it is the required item. If it is, proceed to the next item to fill in. If not, the user needs to be guided. The basis of the guidance is the initial database of error points and the database of solutions established earlier. For example, in a certain step, the customer entered a mobile phone number when he needed to enter an ID number. This will cause the system to report an error, and the system will guide the user to fill in the ID number. Of course, this is just an extremely simple example, which is used for easy understanding. There are often many complex error points in actual operations, which will not be repeated here.

[0044] S3. Prioritize the error points and solutions that meet the preset matching degree; if the user confirms that the problem is solved and correctly proceeds to the next step, execute step S4; if the user's problem is not solved, execute step S5;

[0045] What needs to be understood here is that the preset matching degree is not necessarily a numerical matching degree, but may also be the top few matching degrees, such as the top 3 error points and solutions of the matching degree; the reason for not displaying all of them at the beginning is to facilitate user browsing. Too many display items can easily cause trouble to users and waste time. Generally speaking, the number of displays is derived from a large amount of historical data. When a considerable number of solutions are displayed, the problem in most cases can be solved. In other words, if the problem cannot be solved by displaying the solutions in the above proportion, the number of cases that can be solved when all of them are displayed is relatively small. Therefore, the best way to deal with it is to display a considerable number of solutions. If displaying a considerable number of solutions cannot solve the problem, then it will be optimized in subsequent steps so that the problem can be solved within a considerable number of display items.

[0046] S4. Repeat step S2 after entering the next step; each step is started according to the above process.

[0047] S5. Display multiple matching results in order of matching degree and highlight the matching points of the initial error point database and the data filled in by the user for the user to judge; if the user problem is solved, execute step S6; if the user problem is not solved, execute step S7;

[0048] There are two purposes of highlighting. One is to make it clear to users at a glance. Second, if the highlighted part is not what the user thinks, it can be corrected in time according to user needs, laying the foundation for the update and practicality of the database.

[0049] S6. Based on the fuzzy matching results of the problem, the fuzzy matching optimization is performed again on the data entered by the user so that the user's selection reaches the preset matching degree and the fuzzy matching solution is updated;

[0050] If displaying a considerable number of results fails to solve the problem, but increasing the number of results does solve the problem, then it means that the database is complete, but there is a deviation between the fuzzy matching rules and the required results. The fuzzy matching scheme needs to be reversed based on the required results so that the required data is displayed first.

[0051] S7. Display a questionnaire to the user regarding the incorrect data provided by the user. After the user completes the questionnaire, the user performs a match against the starting point database based on the questionnaire results. If a second matching result is found that is different from the previous matching result, the user proceeds to step S8; if no matching result is found, the user proceeds to step S9.

[0052] When all the solutions still cannot solve the problem, new possibly suitable data can be matched through questionnaires. At this time, it means that there are big problems with the original fuzzy matching, and it is necessary to make a reverse correction based on the data selected after the questionnaire. After the correction, when such problems are encountered in the future, there is no need to go to the questionnaire step, making the system more intelligent.

[0053] S8. Display the second matching result to the user. If the problem is solved, execute step S6; if not, execute step S9;

[0054] S9. Connect to manual guidance, complete the input of data according to manual guidance and successfully enter the next step, establish a new matching result according to the manual solution process, update the initial error point database and the corresponding solution database and execute step S6.

[0055] The solution obtained after manual guidance is analyzed manually and intelligently, and the database and fuzzy matching scheme are updated to make the system more perfect.

[0056] The method of the present invention adopts an artificial intelligence learning method to continuously update the database and fuzzy matching scheme to make the guidance closer to the user's needs, so that the system is continuously improved, and the guidance on user errors is more accurate, saving time and cost.

[0057] In some embodiments of the present invention, step S41 is further included after step S4. According to the fuzzy matching result of solving the problem, the data filled in by the user is fuzzy matched and optimized again so that the user's choice becomes the solution with the highest matching degree, and the fuzzy matching solution is updated.

[0058] Making the result selected by the user the best result is an optimization of system usage. A continuously optimized system can save time in subsequent use.

[0059] In some embodiments of the present invention, in step S5, if the user problem is solved, step S6 is executed; before executing step S6, step S51 is also included. The user is asked whether the matching points between the highlighted initial error point database and the data information filled in by the user are correct. If not, the matching points between the initial error point database and the data information filled in by the user pointed out are compared and analyzed with the originally highlighted matching points, and the matching relationship is re-established based on the matching points pointed out by the user, so that the matching points pointed out by the user are displayed first.

[0060] In some embodiments of the present invention, step S61 is also included after step S6. After the fuzzy matching scheme is updated, a comparison analysis is performed between the initial error point database pointed out by the user and the matching points of the data filled out by the user and the originally highlighted matching points. The matching points pointed out by the user are re-analyzed and fuzzy matching optimization is performed so that the solution selected by the user becomes the solution with the highest matching degree and the fuzzy matching scheme is updated.

[0061] The highlighted points may be the core points of the fuzzy matching, and the user's error correction of the highlighted points is an indirect feedback to the fuzzy matching solution, which is conducive to the improvement of the system.

[0062] In some embodiments of the present invention, if there is a second matching result different from the previous matching result in step S7, before executing step S8, the following step is also included: applying the problem-solving method of the questionnaire to each step of the system operation, establishing new problem points based on the questionnaire and updating the initial error point database.

[0063] When the questionnaire form is enabled, there is a second result that can be solved, which means that there is a solution in the database, but the previous fuzzy matching failed to give a suitable result. This means that there may be parts in the questionnaire that were not included in the previous fuzzy matching, and these parts are quite important. At this time, all fuzzy matching solutions must be revised in combination with the questionnaire.

[0064] In some embodiments of the present invention, the updating of the fuzzy matching solution, the updating of the initial database of error points, and the corresponding solution database in steps S6 to S9 is performed in a label-based manner, as follows:

[0065] (1) storing each error point, fuzzy matching scheme, and corresponding solution in the error point initial database into a tab page, wherein the tab page includes a title, a body, and a tab bar;

[0066] (2) The main body stores error points, fuzzy matching solutions, and corresponding solutions, and stores updated error points, fuzzy matching solutions, and corresponding solutions in the form of revision annotations in the tag column, wherein the revision annotations are in the form of hyperlinks;

[0067] (3) Clicking on the hyperlink of the revision annotation generates a new tab page, which contains the new error points, fuzzy matching schemes, and corresponding solutions after the revision annotation;

[0068] (4) The generated new error points, fuzzy matching schemes, and corresponding solutions are stored in the main body of a new tab page, with the title set to a subpage of the original tab page title. The subpages are marked with features for distinguishing each subpage, including: time, reason for revision, and number of users targeted by the revision;

[0069] (5) Repeat step (2) on each subpage.

[0070] The advantage of this storage method is that each version is stored, and related revisions are stored through revision mode, so that related items are linked together. After finding a certain tab page, the pages before and after it can be obtained by clicking the hyperlink to obtain the complete page, so that the revision record is saved continuously.

[0071] In some embodiments of the present invention, after the user completes the use of the new system, a statistical analysis is performed on the data filled in by the user to find defects in the data filled in by the user in the specific business. If the defective item does not reach the set standard for affecting normal operation, the defective item continues to be recorded; if the defective item has reached the warning value for affecting normal operation, the system operation process is re-specified based on the defective item analysis and the defective item solution is obtained to obtain a new process, and steps S1-S9 are re-executed according to the new process.

[0072] In some embodiments of the present invention, whether there are matters requiring approval in the monitoring system, if so, the user is reminded through an early warning channel to complete the approval process of the business document.

[0073] In some embodiments of the present invention, the warning channels include text messages, enterprise WeChat and pop-up services.

[0074] An AI-based system operation intelligent assistant system is used to implement the above assistant method, including:

[0075] Operating system, including operating procedures and operation tutorials;

[0076] Initial database: includes an initial database of error points and a corresponding solution database, used to store initial error points and corresponding solutions sorted out based on the error points in the initial error point database;

[0077] The judgment module is used to judge whether the user operation process is correct;

[0078] Guidance module: if the user operates incorrectly, it will analyze the error and provide solutions for the user to choose;

[0079] The updating module is configured to update the initial error point database and the corresponding solution database according to the user's feedback in the guidance to form a new database.

[0080] It should be noted that the above embodiments can be freely combined as needed. The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that the present invention is susceptible to various modifications and variations. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.

Claims

1. An AI-based system operation intelligent assistant method, characterized in that: The steps include: S1. After establishing a new system, establish operational procedures. Based on the operational procedures, predict problems for the new system and conduct system usage investigation and testing. Summarize the predicted and investigated error points to create an initial error database. Based on the error points in the initial error database, create an initial database of corresponding solutions. S2 determines whether the data filled in by the user is correct. If correct, proceed to the next step. If wrong, the data filled in by the user is fuzzy matched with the initial database of error points; if there is a fuzzy match result, proceed to step S3; If there is no matching result, execute step S7; S3. Prioritize the error points and solutions that meet the preset matching degree; if the user confirms that the problem is solved and correctly proceeds to the next step, execute step S4; if the user's problem is not solved, execute step S5; S4. Repeat step S2 after entering the next step; S5. Display multiple matching results in order of matching degree and highlight the matching points of the initial database and the user-filled data for the user to judge; If the user problem is solved, go to step S6; if the user problem is not solved, go to step S7; S6. Based on the fuzzy matching results of the problem, the fuzzy matching optimization is performed again on the data entered by the user so that the user's selection reaches the preset matching degree and the fuzzy matching solution is updated; S7. Display the questionnaire for the user's data error to the user. After the user completes the questionnaire, the results of the questionnaire are matched in the starting point database. If there is a second matching result that is different from the previous matching result, step S8 is executed; If there is no matching result, execute step S9; S8. Display the second matching result to the user. If the problem is solved, execute step S6; If the problem is not solved, proceed to step S9; S9. Connect to manual guidance, complete the data input according to manual guidance and successfully enter the next step, establish a new matching result according to the manual solution process, update the initial database of error points and the corresponding solution database and execute step S6.

2. The AI-based system operation intelligent assistant method according to claim 1, characterized in that: Step S4 also includes step S41. According to the fuzzy matching result of solving the problem, the data filled in by the user is fuzzy matched and optimized again so that the user's choice becomes the solution with the highest matching degree, and the fuzzy matching solution is updated.

3. The AI-based system operation intelligent assistant method according to claim 1, characterized in that: In step S5, if the user problem is solved, step S6 is executed; before executing step S6, step S51 is also included. The user is asked whether the matching points of the initial database with the highlighted error points and the data information filled in by the user are correct. If not, the matching points of the initial database with the error points pointed out by the user and the data information filled in by the user are compared and analyzed with the originally highlighted matching points, and the matching relationship is re-established based on the matching points pointed out by the user, so that the matching points pointed out by the user are displayed first.

4. The AI-based system operation intelligent assistant method according to claim 3, characterized in that: Step S6 also includes step S61. After updating the fuzzy matching solution, the matching points of the initial database and the data filled in by the user are compared and analyzed with the originally highlighted matching points based on the error points pointed out by the user. The fuzzy matching is re-analyzed and optimized according to the matching points pointed out by the user so that the solution selected by the user becomes the solution with the highest matching degree and the fuzzy matching solution is updated.

5. The AI-based system operation intelligent assistant method according to claim 1, characterized in that: If there is a second matching result different from the previous matching result in step S7, before executing step S8, the following steps are also included: applying the problem-solving method of the questionnaire to each step of the system operation, establishing new problem points based on the questionnaire and updating the initial database of error points.

6. The AI-based system operation intelligent assistant method according to claim 1, characterized in that: In steps S6 to S9, the fuzzy matching scheme, the initial database of error points, and the corresponding solution database are updated using a label method. The steps are as follows: (1) storing each error point, fuzzy matching scheme and corresponding solution in the error point initial database into a tab page, wherein the tab page includes a title, a body and a tab bar; (2) The main body stores the error points, fuzzy matching solutions and corresponding solutions, and stores the updated error points, fuzzy matching solutions and corresponding solutions in the form of revision annotations in the tag column, wherein the revision annotations are in the form of hyperlinks; (3) Click the hyperlink of the revision annotation to generate a new tab page, which contains the new error points, fuzzy matching schemes and corresponding solutions after the revision annotation; (4) The generated new error points, fuzzy matching schemes and corresponding solutions are stored in the main body of a new tab page, with the title set as a subpage of the original tab page title. The subpages are marked with features for distinguishing each subpage, including: time, reason for revision and the number of users targeted by the revision; (5) Repeat step (2) on each subpage.

7. The AI-based system operation intelligent assistant method according to claim 1, characterized in that: After the user completes the use of the new system, the user's data is statistically analyzed to identify any defects in the data used in the new system in specific business operations. If the defect does not meet the set standard for affecting normal operation, the defect will continue to be recorded. If the defect item has reached the warning value that affects normal operation, the system operation process is re-specified to obtain a new process based on the defect item analysis and the defect item solution, and steps S1-S9 are re-executed according to the new process.

8. The AI-based system operation intelligent assistant method according to claim 1, characterized in that: Monitor whether there are any matters that require approval in the system. If so, remind the user through the early warning channel to complete the approval process of the business document.

9. The AI-based system operation intelligent assistant method according to claim 8, characterized in that: The warning channels include SMS, enterprise WeChat and pop-up services.

10. An AI-based system operation intelligent assistant system, characterized in that: The assistant method for completing any one of claims 1 to 9 comprises: Operating system, including operating procedures and operation tutorials; Initial database: includes an initial database of error points and a corresponding solution database, used to store initial error points and corresponding solutions sorted out based on the error points in the initial database of error points; The judgment module is used to judge whether the user operation process is correct; Guidance module: if the user operates incorrectly, it will analyze the error and provide solutions for the user to choose; The updating module is configured to update the initial database of error points and the corresponding solution database to form a new database according to the user's feedback in the guidance.