A data maintenance method, device, equipment and medium

Abstract

The present invention discloses a data maintenance method, device, equipment and medium. The data maintenance method includes: based on natural language processing technology and a large language model, processing the historical data maintenance form of the target time window to obtain the historical data to be matched; matching the target data maintenance form currently submitted by the user with the historical data to be matched, performing content similarity matching and SQL similarity matching to determine the system functions to be improved; and reporting or releasing the rectification plan data of the system functions to be improved according to the data maintenance reporting rules. The technical solution of the embodiment of the present invention can reduce the labor cost of improving the system functions and improve the effect of improving the system functions.

Description

Technical Field

[0001] The present invention relates to the field of data maintenance technology, and in particular to a data maintenance method, device, equipment and medium. Background Art

[0002] In the early stages of the construction of the Xinchuang system (information innovation system), there were problems such as incomplete functions and inadequate construction, resulting in most systems generating a large number of data maintenance orders every month and repeated data maintenance.

[0003] In order to improve system functions and reduce operation and maintenance costs, relevant operation and maintenance personnel conduct statistical analysis every month to find out where the system needs to be improved. According to statistics, individual systems generate about 500 data maintenance orders on average each month. Faced with a large number of documents, operation and maintenance personnel need to mark and classify data maintenance orders for different reasons, and only after manual statistical analysis can they find out the functions that each system currently does not have but frequently needs to use, and use this as feedback to developers to improve the construction of the system. However, manual statistical analysis of data maintenance orders has problems such as high work repetitiveness and easy omissions in manual statistics, and requires a lot of manpower every month. Summary of the invention

[0004] The present invention provides a data maintenance method, device, equipment and medium to solve the problems of high labor cost and poor system function improvement effect in the current system function improvement based on manual statistical analysis.

[0005] According to one aspect of the present invention, there is provided a data maintenance method, comprising:

[0006] Based on natural language processing technology and large language models, the historical data maintenance order of the target time window is processed to obtain the historical data to be matched;

[0007] Perform content similarity matching and SQL similarity matching on the target data maintenance form currently submitted by the user and the historical data to be matched, and determine the functions to be improved in the system;

[0008] According to the data maintenance and reporting rules, the rectification plan data for the system functions to be improved shall be reported or released.

[0009] According to another aspect of the present invention, there is provided a data maintenance device, comprising:

[0010] The historical data maintenance form processing module is used to process the historical data maintenance form of the target time window based on natural language processing technology and a large language model to obtain historical data to be matched;

[0011] The similarity matching module is used to perform content similarity matching and SQL similarity matching on the target data maintenance form currently submitted by the user and the historical data to be matched, and determine the functions to be improved in the system;

[0012] The system function improvement module is used to report or release the rectification plan data of the system functions to be improved according to the data maintenance reporting rules.

[0013] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0014] at least one processor; and

[0015] a memory communicatively connected to the at least one processor; wherein,

[0016] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the data maintenance method described in any embodiment of the present invention.

[0017] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the data maintenance method described in any embodiment of the present invention when executed.

[0018] The technical solution of the embodiment of the present invention processes the historical data maintenance form of the target time window based on natural language processing technology and a large language model to obtain historical data to be matched, thereby matching the target data maintenance form currently submitted by the user with the historical data to be matched for content similarity and SQL similarity, determining the system functions to be improved, and then reporting or releasing the rectification plan data of the system functions to be improved according to the data maintenance reporting rules. In this solution, by performing two types of similarity matching between the target data maintenance form and the historical data to be matched, the frequently occurring similar data maintenance forms are quickly and accurately screened out, and then the system function defects are automatically analyzed based on the similar data maintenance forms, that is, the system functions to be improved are determined, and based on the data maintenance reporting rules, the rectification plan data of the system functions to be improved are differentiated and orderly handled, which solves the current problems of high labor costs and poor system function improvement effects in the system function improvement based on manual statistical analysis, and can reduce the labor costs of system function improvement and improve the system function improvement effects.

[0019] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. 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 creative work.

[0021] Figure 1 A flowchart of a data maintenance method provided in Embodiment 1 of the present invention;

[0022] Figure 2 A flowchart of a data maintenance method provided in Embodiment 2 of the present invention;

[0023] Figure 3 A schematic diagram of the structure of a data maintenance device provided in Embodiment 3 of the present invention;

[0024] Figure 4 A schematic diagram of the structure of an electronic device that can be used to implement an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0025] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0026] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0027] Embodiment 1

[0028] Figure 1This is a flow chart of a data maintenance method provided in the first embodiment of the present invention. This embodiment can be applied to the situation of automatically determining the functions required to be improved in the information innovation system. The method can be executed by a data maintenance device, which can be implemented in the form of hardware and / or software. The data maintenance device can be configured in electronic devices such as servers. Figure 1 As shown, the method includes:

[0029] Step 110: Based on natural language processing technology and a large language model, the historical data maintenance sheet of the target time window is processed to obtain historical data to be matched.

[0030] Among them, natural language processing technology is a technology that uses computer technology to analyze, understand and process natural language with language as the object, namely NLP (Natural Language Processing). A large language model refers to an advanced artificial intelligence algorithm trained on a large amount of data. The target time window can be a pre-set time window, and the corresponding duration of the target time window can be set based on user needs. The historical data maintenance form can be a data maintenance form for a period of time in the past that needs to be compared and analyzed with the currently submitted data maintenance form. The historical data to be matched can be the historical data to be matched obtained by analyzing and processing the historical data maintenance form based on natural language processing technology and a large language model.

[0031] In an embodiment of the present invention, a scheduled task can be set to collect and process historical data maintenance orders. According to the timing period of the scheduled task, the historical data maintenance orders of the target time window are collected, and then natural language processing technology and a large language model are used to perform statistical analysis on the historical data maintenance orders to obtain historical data to be matched.

[0032] Exemplarily, the target time window may be set to 6 months before the aforementioned scheduled task is triggered this time, and the timing period of the scheduled task may be set as needed, such as 1 day, 3 days, or 1 week.

[0033] Step 120: Perform content similarity matching and SQL similarity matching on the target data maintenance form currently submitted by the user and the historical data to be matched, and determine the functions to be improved in the system.

[0034] Among them, the target data maintenance form can be the data maintenance form currently submitted by the user. Content similarity matching can be an operation of performing similarity analysis on the form contents of different data maintenance forms. SQL similarity matching can be an operation of performing similarity analysis on different SQLs. Exemplarily, SQL similarity matching can be used to perform similarity analysis on different SQLs from the dimensions of database, table, and field. The system functions to be improved can be system functions that cause operation and maintenance personnel to frequently perform data maintenance (exceeding the preset maintenance frequency) due to functional defects in the information creation system. Exemplarily, the system functions to be improved can include but are not limited to adding new data.

[0035] The target data maintenance form and the historical data maintenance form have the same work order format. The content for operation and maintenance personnel to fill in the work order includes application platform type (such as SaaS platform), system name, database name, execution user, backup SQL, execution SQL, task description, and execution type. The task description is the cause and purpose of system maintenance filled in by the operation and maintenance personnel, and the minimum number of words in the task description is limited to ensure that the operation and maintenance personnel can fill in key information normally and reserve space for analyzing documents.

[0036] In an embodiment of the present invention, the target data maintenance form currently submitted by the user and the historical data to be matched can be matched for content similarity, and the target data maintenance form currently submitted by the user and the corresponding SQL of the historical data to be matched can be matched for SQL similarity, so as to parse the data maintenance form that meets the preset similarity requirements, and then the system function corresponding to the data maintenance form that meets the preset similarity requirements can be used as the system function to be improved.

[0037] Step 130: According to the data maintenance reporting rules, the rectification plan data of the system functions to be improved is reported or released.

[0038] The data maintenance reporting rules may be pre-set handling rules for the system functions to be improved. The rectification plan data may be data describing the system functions to be improved, which is used for approval by the person in charge of the system. Exemplarily, the rectification plan data may include a functional description of the system functions to be improved, the time when the rectification plan data is generated, the rectification cycle, and the information of the receiving personnel.

[0039] In an embodiment of the present invention, it is possible to determine whether the rectification plan data corresponding to the system's to-be-perfected function has been reported based on the data maintenance and reporting rules. If it has not been reported, the rectification plan data will be reported, or a corresponding work order will be generated based on the rectification plan data and sent to the person in charge of the system. If similar rectification plan data (system functions to be perfected that belong to the same system as the rectification plan data determined this time and have similar functions) has been reported within a preset period, it indicates that the system's to-be-perfected function is being perfected, and the current rectification plan data is released, that is, the current rectification plan data is not reported. If the time when similar rectification plan data was previously reported is greater than the preset period, it indicates that the system's to-be-perfected function has not been optimized within the specified time, and the data maintenance order corresponding to the current rectification plan data is locked, and the rectification plan data and similar rectification plan data are integrated and summarized and reported to the superior. Through this solution, the person in charge of the relevant system can be urged to complete the missing functions and improve the construction of the system.

[0040] The technical solution of the embodiment of the present invention processes the historical data maintenance form of the target time window based on natural language processing technology and a large language model to obtain historical data to be matched, thereby matching the target data maintenance form currently submitted by the user with the historical data to be matched for content similarity and SQL similarity, determining the system functions to be improved, and then reporting or releasing the rectification plan data of the system functions to be improved according to the data maintenance reporting rules. In this solution, by performing two types of similarity matching between the target data maintenance form and the historical data to be matched, the frequently occurring similar data maintenance forms are quickly and accurately screened out, and then the system function defects are automatically analyzed based on the similar data maintenance forms, that is, the system functions to be improved are determined, and based on the data maintenance reporting rules, the rectification plan data of the system functions to be improved are differentiated and orderly handled, which solves the current problems of high labor costs and poor system function improvement effects in the system function improvement based on manual statistical analysis, and can reduce the labor costs of system function improvement and improve the system function improvement effects.

[0041] Embodiment 2

[0042] Figure 2 This is a flowchart of a data maintenance method provided in the second embodiment of the present invention. This embodiment is specific based on the above embodiment and provides a specific optional implementation method for processing the historical data maintenance list of the target time window based on natural language processing technology and a large language model to obtain the historical data to be matched. Figure 2 As shown, the method includes:

[0043] Step 210: Based on natural language processing technology, the historical data maintenance orders of the target time window are traversed and matched to determine similar historical data maintenance orders.

[0044] The similar historical data maintenance order may be a historical data maintenance order that meets a similarity threshold in the same target time window.

[0045] In an embodiment of the present invention, historical data maintenance sheets of the target time window can be compared in pairs based on natural language processing technology, that is, historical data maintenance sheets of the target time window are traversed and matched in pairs to obtain similar historical data maintenance sheets that meet the similarity threshold.

[0046] Optionally, based on Python's spacy library, historical data maintenance orders of the target time window can be traversed and matched to determine similar historical data maintenance orders.

[0047] For example, first prepare the data sample set, then perform model training, and after training, you can get the corresponding model package. After using spacy to load the language and the trained model package, you can put the historical data maintenance sheet into spacy. At this time, the characters (Token.vector), words (Span.vector), and sentences (Doc.vector) in the text will be assigned corresponding vectors. Since Chinese characters have multiple meanings and polyphonic characters, in the actual processing process, the similarity comparison of two texts is mainly determined by comparing sentence vectors, followed by word vectors, first overall and then locally. Each sentence has a grammatical structure, so through dependency syntactic analysis, the structure of the sentence and the dependency relationship between words can be extracted. Based on these dependencies and the vector of the entire sentence, directly use the similarity() method provided by spacy to compare whether the two sentences are similar and obtain the similarity data of the two sentences.

[0048] The internal meaning of the text processed by the spacy pipeline will be represented as a floating point array, also known as a tensor. This allows spacy to make reasonable guesses based on the meaning of the surrounding words and to make connections based on the context. Suppose there are three sentences: Maine can meow; Maine can wag its tail at people; Maine is a loyal companion of humans, and the American place name "Maine" is included in different contexts. Although the model does not know the existence of "Maine cat" at the beginning, when Maine and cat are matched at the beginning, the similarity is only 0.1. After adding some habits of cats to the composed sentences, the model recognizes that the similarity can reach 0.4. Using the feature that spacy can make reasonable guesses based on the meaning of the surrounding words, the vectors of the words can be corrected, and the vectors between similar words will be closer, thereby improving the accuracy of the overall judgment.

[0049] In summary, first directly match the similarity of the two historical data maintenance orders in the whole sentence, that is, calculate the vector of the whole sentence, and get the first segment of similarity data. Then parse the grammatical structure of the two historical data maintenance orders respectively, and perform structural vector calculation according to the dependency relationship between each word to obtain the second segment of similarity data. Finally, based on the context, make some relatively uncommon proprietary words, such as fee control, red rush, blue rush, etc., understand the semantics, that is, correct the values ​​of the floating point arrays of these words, and then match them to get the third segment of similarity data. Finally, assign weights to the three segments of similarity data, calculate their weighted average, and get their final similarity. The weight ratio of the three will be fine-tuned according to the actual situation. It has been verified that when the weight ratio is 4.13:2.06:3.81, it can handle most situations and find data maintenance orders with similar reasons.

[0050] Step 220: Classify similar historical data maintenance orders according to the large language model to determine historical data to be matched.

[0051] In the embodiment of the present invention, a large language model can be used to identify the system type to which similar historical data maintenance orders belong, and to classify the data maintenance orders belonging to the same system to obtain historical data to be matched.

[0052] It should be noted that the embodiment of the present invention can also first use a large language model to classify the historical data maintenance sheets of the target time window according to the system to which they belong, and then use natural language processing technology to traverse and match the classified data maintenance sheets to obtain historical data to be matched.

[0053] Step 230: Perform content similarity matching and SQL similarity matching on the target data maintenance form currently submitted by the user and the historical data to be matched, and determine the functions to be improved in the system.

[0054] In an optional embodiment of the present invention, the target data maintenance form currently submitted by the user is matched with the historical data to be matched, and content similarity matching and SQL similarity matching are performed to determine the system functions to be improved, which may include: matching the target data maintenance form currently submitted by the user with the historical data to be matched, and obtaining SQL data to be matched; performing SQL similarity matching on the SQL data to be matched to determine the repeated triggering of the data maintenance form, and determining the system functions to be improved based on the repeated triggering of the data maintenance form.

[0055] The SQL data to be matched may be the SQL of similar data maintenance orders determined after content similarity matching between the target data maintenance order and the historical data to be matched. The repeatedly triggered data maintenance order may be the repeatedly triggered similar data maintenance order determined based on the SQL data to be matched.

[0056] In an embodiment of the present invention, the target data maintenance form currently submitted by the user can be matched with the historical data to be matched for content similarity, a data maintenance form that reaches a certain similarity can be determined, and the SQL of the aforementioned data maintenance form that reaches a certain similarity can be obtained, that is, the SQL data to be matched can be obtained, and SQL similarity matching can be further performed on the SQL data to be matched, and the data maintenance form with a certain SQL similarity can be used as a repeated trigger data maintenance form, and based on the task description corresponding to the repeated trigger data maintenance form, the system function to be improved can be determined.

[0057] In an optional embodiment of the present invention, the target data maintenance form currently submitted by the user is matched with the historical data to be matched for content similarity to obtain SQL data to be matched, which may include: based on natural language processing technology and a large language model, matching the target data maintenance form currently submitted by the user with the historical data to be matched for content similarity, and determining a target historical data maintenance form in the historical data to be matched whose content similarity with the target data maintenance form is greater than or equal to a content similarity threshold; and determining SQL data to be matched based on the target historical data maintenance form and the target data maintenance form.

[0058] The content similarity threshold can be used to determine whether the sentence contents of different data maintenance sheets are similar. The target historical data maintenance sheet can be a historical data maintenance sheet whose similarity with the target data maintenance sheet is greater than or equal to the content similarity threshold.

[0059] In an embodiment of the present invention, the target data maintenance form currently submitted by the user can be systematically classified based on a large language model, and based on natural language processing technology, the target data maintenance form and the historical data to be matched are traversed and matched based on text content, that is, content similarity matching is performed, and the target historical data maintenance form whose content similarity with the target data maintenance form is greater than or equal to the content similarity threshold in the historical data to be matched is determined, and the SQL associated with the target historical data maintenance form and the target data maintenance form respectively is further determined to obtain the SQL data to be matched.

[0060] Optionally, Python's spacy library can be used to perform content similarity matching between the target data maintenance sheet currently submitted by the user and the historical data to be matched.

[0061] In an optional embodiment of the present invention, SQL similarity matching is performed on the SQL data to be matched, a repeated triggering data maintenance form is determined, and based on the repeated triggering data maintenance form, the system function to be improved is determined, which may include: performing SQL similarity matching on the SQL data to be matched to obtain a repeated triggering data maintenance form with an SQL similarity greater than or equal to an SQL similarity threshold; parsing the task description field in the repeated triggering data maintenance form to obtain the system function to be improved.

[0062] The SQL similarity threshold can be used to determine whether different SQLs are similar. The task description field can be a field of the task description in the data maintenance form.

[0063] In an embodiment of the present invention, SQL data to be matched can be subjected to SQL similarity matching to obtain SQL in the SQL data to be matched whose similarity is greater than or equal to the SQL similarity threshold, i.e., similar SQL. Then, the data maintenance form of the similar SQL can be used as a repeated triggering data maintenance form, and the task description field of the repeated triggering data maintenance form can be parsed to use the system function corresponding to the task description field as the system function to be improved.

[0064] In an optional embodiment of the present invention, determining SQL data to be matched based on the target historical data maintenance sheet and the target data maintenance sheet may include: when the number of data sheets of the target historical data maintenance sheet is greater than or equal to a preset data sheet number threshold, querying the first SQL associated statement of the target historical data maintenance sheet and the second SQL associated statement of the target data maintenance sheet; and using the first SQL associated statement and the second SQL associated statement as SQL data to be matched.

[0065] The data sheet quantity threshold may be a preset number of data sheets. The first SQL association statement may be an SQL corresponding to the target historical data maintenance sheet. The second SQL association statement may be an SQL corresponding to the target data maintenance sheet.

[0066] In an embodiment of the present invention, the number of data sheets of the target historical data maintenance sheet can be counted, and the number of data sheets of the target historical data maintenance sheet can be compared with a preset data sheet number threshold. Therefore, when the number of data sheets of the target historical data maintenance sheet is greater than or equal to the preset data sheet number threshold, the first SQL associated statement of the target historical data maintenance sheet and the second SQL associated statement of the target data maintenance sheet are queried from the SQL database storing the data maintenance sheet, and the first SQL associated statement and the second SQL associated statement are used as SQL data to be matched.

[0067] Step 240: According to the data maintenance reporting rules, the rectification plan data of the system functions to be improved is reported or released.

[0068] In an optional embodiment of the present invention, the rectification plan data of the system functions to be improved are reported or released according to the data maintenance reporting rules, which may include: determining the rectification plan data of the system functions to be improved within the data maintenance rectification cycle according to the data maintenance reporting rules; reporting or releasing the rectification plan data according to the number of times similar rectification plan data are generated within the data maintenance rectification cycle.

[0069] Among them, the data maintenance and rectification cycle can be the data maintenance cycle specified in the data maintenance reporting rules.

[0070] In an embodiment of the present invention, the data maintenance rectification cycle can be determined based on the data maintenance reporting rules, and the rectification plan data of the system functions to be improved within the data maintenance rectification cycle can be obtained. Further, based on the number of times similar rectification plan data are generated within the data maintenance rectification cycle, the rectification plan data can be automatically reported or released in accordance with the data maintenance reporting rules.

[0071] In a specific example, because the page of system A lacks the function of "adding a new piece of data", the operation and maintenance personnel have to perform data maintenance 8 times. It is necessary to add a "add data" entry to provide users with operations on the page. Configure a scheduled task to collect and process historical data maintenance orders, and store the scheduled analysis results in the database for subsequent process calls. When the user submits the target data maintenance order, the user will be asked to fill in the task description on the page. If the reason is normal (meets the word limit), the user can click the verification button configured on the page to connect to the aforementioned database to perform SQL syntax verification, SQL execution plan, semantic analysis and other verification operations. Among them, the semantic analysis operation will extract the results of the previous scheduled analysis and storage, and perform similarity matching with the user's reasons. The specific similarity matching process is as follows:

[0072] First determine the system that needs data maintenance, then extract the historical data to be matched from the database, and perform semantic analysis and content similarity matching on the extracted historical data to be matched with the "task description" in the target data maintenance form. When the similarity between the two matches reaches the content similarity threshold (90%), it will be determined whether it has been triggered repeatedly (5 times). If the similarity does not reach 90% or the number of triggers is less than 5 times, it is determined as uncommon data maintenance and released.

[0073] Further, determine the target historical data maintenance sheet with a content similarity of no less than 90% and a data maintenance reason repeated 5 or more times, and then query the SQL data to be matched based on the target data maintenance sheet and the target historical data maintenance sheet, so as to determine whether the degree of duplication of the database, table, and field involved in the SQL data to be matched reaches (SQL similarity threshold) 80%. When all the above conditions are met, it means that the system function construction is not perfect, and frequent data maintenance is required to realize the functions required by users, and rectification should be ordered, and then the subsequent step of generating a rectification plan will be entered.

[0074] Determine whether a rectification plan has been generated for the system manager within the data maintenance and rectification cycle (within 30 days). If not, record the time of the trigger and related information, and generate a corresponding work order for the system manager, requesting rectification, and prompting the system to generate a rectification plan for him, and the subsequent process will proceed as usual. If similar rectification plan data has been generated before, further determine whether it was generated within the data maintenance and rectification cycle. If so, it is considered that the function is being improved and released, and the subsequent process will proceed as usual. If the time from the last time similar rectification plan data was generated is greater than the data maintenance and rectification cycle, it means that the developer has not completed the optimization within the specified time. At this time, the corresponding data maintenance order will be locked, and the superior leader needs to approve it, otherwise the data maintenance will not be performed. The process of this plan is aimed at improving the construction of system functions. Avoid the situation in which users cannot operate through the page due to the lack of a certain function during the system construction process, and can only frequently use data maintenance to achieve their goals, which will lead to problems such as low office efficiency, increased operation and maintenance costs, and repeated work. This plan uses statistical analysis to find similar data maintenance orders and urges relevant systems to rectify, which can fill the system's shortcomings faster and solve the above problems.

[0075] This solution introduces natural language processing technology and large language models for analysis, summarizes the data maintenance that needs to be repeated due to the lack of a certain function, and generates corresponding rectification plan data to order the relevant system leaders to improve the system construction, thereby reducing the amount of data maintenance, that is, reducing the operation and maintenance labor costs and hardware costs, and indirectly improving office efficiency. And through this solution, it can replace the manual analysis of massive documents, solve the problems of omissions, high repetition, high labor costs, and high subjective judgment factors that are prone to occur in manual statistics, and can realize the tracking of the entire life cycle of the rectification plan, and confirm whether the system has supplemented the missing functions as required, and assist in improving the construction of the system through data maintenance.

[0076] The technical solution of the embodiment of the present invention, through the natural language processing technology, traverses and matches the historical data maintenance orders of the target time window, determines the similar historical data maintenance orders, and then classifies the similar historical data maintenance orders according to the large language model, determines the historical data to be matched, and then matches the target data maintenance order currently submitted by the user with the historical data to be matched, performs content similarity matching and SQL similarity matching, determines the system functions to be improved, and reports or releases the rectification plan data of the system functions to be improved according to the data maintenance reporting rules. In this scheme, by performing two types of similarity matching between the target data maintenance order and the historical data to be matched, the frequently occurring similar data maintenance orders are quickly and accurately screened out, and then the system function defects are automatically analyzed based on the similar data maintenance orders, that is, the system functions to be improved are determined, and based on the data maintenance reporting rules, the rectification plan data of the system functions to be improved are differentiated and orderly handled, which solves the current problems of high labor costs and poor system function improvement effects in the system function improvement based on manual statistical analysis, can reduce the labor costs of system function improvement, and improve the system function improvement effect.

[0077] Embodiment 3

[0078] Figure 3 This is a schematic diagram of the structure of a data maintenance device provided in Embodiment 3 of the present invention. Figure 3 As shown, the device comprises:

[0079] The historical data maintenance sheet processing module 310 is used to process the historical data maintenance sheet of the target time window based on natural language processing technology and a large language model to obtain historical data to be matched.

[0080] The similarity matching module 320 is used to perform content similarity matching and structured query language SQL similarity matching on the target data maintenance form currently submitted by the user and the historical data to be matched, so as to determine the functions to be improved in the system.

[0081] The system function improvement module 330 is used to report or release the rectification plan data of the system functions to be improved according to the data maintenance reporting rules.

[0082] The technical solution of the embodiment of the present invention processes the historical data maintenance form of the target time window based on natural language processing technology and a large language model to obtain historical data to be matched, thereby matching the target data maintenance form currently submitted by the user with the historical data to be matched for content similarity and SQL similarity, determining the system functions to be improved, and then reporting or releasing the rectification plan data of the system functions to be improved according to the data maintenance reporting rules. In this solution, by performing two types of similarity matching between the target data maintenance form and the historical data to be matched, the frequently occurring similar data maintenance forms are quickly and accurately screened out, and then the system function defects are automatically analyzed based on the similar data maintenance forms, that is, the system functions to be improved are determined, and based on the data maintenance reporting rules, the rectification plan data of the system functions to be improved are differentiated and orderly handled, which solves the current problems of high labor costs and poor system function improvement effects in the system function improvement based on manual statistical analysis, and can reduce the labor costs of system function improvement and improve the system function improvement effects.

[0083] Optionally, the historical data maintenance form processing module 310 is specifically used to traverse and match the historical data maintenance forms of the target time window based on the natural language processing technology to determine similar historical data maintenance forms; and classify the similar historical data maintenance forms according to the large language model to determine the historical data to be matched.

[0084] Optionally, the similarity matching module 320 includes a first matching unit and a second matching unit; the first matching unit is used to perform content similarity matching on the target data maintenance form currently submitted by the user and the historical data to be matched to obtain SQL data to be matched. The second matching unit is used to perform SQL similarity matching on the SQL data to be matched to determine the repeated triggering data maintenance form, and determine the system function to be improved based on the repeated triggering data maintenance form.

[0085] Optionally, the first matching unit is specifically used to perform content similarity matching between the target data maintenance form currently submitted by the user and the historical data to be matched based on the natural language processing technology and the large language model, and determine a target historical data maintenance form in the historical data to be matched whose content similarity with the target data maintenance form is greater than or equal to a content similarity threshold; and determine the SQL data to be matched based on the target historical data maintenance form and the target data maintenance form.

[0086] Optionally, the second matching unit is specifically used to perform SQL similarity matching on the SQL data to be matched to obtain the repeated triggering data maintenance form with SQL similarity greater than or equal to the SQL similarity threshold; and parse the task description field in the repeated triggering data maintenance form to obtain the system function to be improved.

[0087] Optionally, the first matching unit is specifically used to query the first SQL associated statement of the target historical data maintenance sheet and the second SQL associated statement of the target data maintenance sheet when the number of data sheets of the target historical data maintenance sheet is greater than or equal to a preset data sheet number threshold; and use the first SQL associated statement and the second SQL associated statement as the SQL data to be matched.

[0088] Optionally, the system function improvement module 330 is specifically used to determine the rectification plan data of the system functions to be improved within the data maintenance rectification cycle according to the data maintenance reporting rules; and report or release the rectification plan data according to the number of times similar rectification plan data are generated within the data maintenance rectification cycle.

[0089] The data maintenance device provided in the embodiment of the present invention can execute the data maintenance method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0090] Embodiment 4

[0091] Figure 4 A schematic diagram of an electronic device that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0092] like Figure 4 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. The input / output (I / O) interface 15 is also connected to the bus 14.

[0093] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0094] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as a data maintenance method.

[0095] In some embodiments, the data maintenance method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the data maintenance method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to perform the data maintenance method in any other appropriate manner (e.g., by means of firmware).

[0096] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0097] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0098] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, device, or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0099] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0100] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0101] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0102] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0103] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A data maintenance method, characterized in that: include: Based on natural language processing technology and large language models, the historical data maintenance order of the target time window is processed to obtain the historical data to be matched; The target data maintenance form currently submitted by the user is matched with the historical data to be matched, and content similarity matching and structured query language SQL similarity matching are performed to determine the functions to be improved in the system; According to the data maintenance reporting rules, report or release the rectification plan data of the system functions to be improved; Based on natural language processing technology and large language models, the historical data maintenance order of the target time window is processed to obtain historical data to be matched, including: Based on the natural language processing technology, the historical data maintenance orders of the target time window are traversed and matched to determine similar historical data maintenance orders; according to the large language model, the similar historical data maintenance orders are classified to determine the historical data to be matched; The target data maintenance form currently submitted by the user is matched with the historical data to be matched, and content similarity matching and SQL similarity matching are performed to determine the functions to be improved in the system, including: The target data maintenance form currently submitted by the user is matched with the historical data to be matched for content similarity to obtain SQL data to be matched; SQL similarity matching is performed on the SQL data to be matched to determine the repeated triggering data maintenance form, and based on the repeated triggering data maintenance form, the system function to be improved is determined.

2. The method according to claim 1, characterized in that The target data maintenance form currently submitted by the user is matched with the historical data to be matched, and the SQL data to be matched is obtained, including: Based on the natural language processing technology and the large language model, the target data maintenance form currently submitted by the user is matched with the historical data to be matched for content similarity, and a target historical data maintenance form whose content similarity with the target data maintenance form is greater than or equal to a content similarity threshold value is determined in the historical data to be matched; The SQL data to be matched is determined according to the target historical data maintenance sheet and the target data maintenance sheet.

3. The method according to claim 1, characterized in that Performing SQL similarity matching on the SQL data to be matched, determining a repeated triggering data maintenance sheet, and determining the system function to be improved according to the repeated triggering data maintenance sheet, including: Perform SQL similarity matching on the SQL data to be matched, and obtain the repeated trigger data maintenance order whose SQL similarity is greater than or equal to the SQL similarity threshold; The task description field in the repeated trigger data maintenance form is parsed to obtain the functions to be improved in the system.

4. The method according to claim 2, characterized in that: Determining the SQL data to be matched according to the target historical data maintenance sheet and the target data maintenance sheet includes: When the number of data sheets of the target historical data maintenance sheet is greater than or equal to a preset data sheet number threshold, querying a first SQL associated statement of the target historical data maintenance sheet and a second SQL associated statement of the target data maintenance sheet; The first SQL associated statement and the second SQL associated statement are used as the SQL data to be matched.

5. The method according to claim 1, characterized in that According to the data maintenance reporting rules, report or release the rectification plan data of the system functions to be improved, including: According to the data maintenance reporting rules, determine the rectification plan data of the system functions to be improved within the data maintenance rectification cycle; The rectification plan data is reported or released according to the number of times similar rectification plan data is generated within the data maintenance and rectification cycle.

6. A data maintenance device, characterized in that: include: The historical data maintenance form processing module is used to process the historical data maintenance form of the target time window based on natural language processing technology and a large language model to obtain historical data to be matched; A similarity matching module is used to perform content similarity matching and SQL similarity matching on the target data maintenance form currently submitted by the user and the historical data to be matched, so as to determine the functions to be improved in the system; The system function improvement module is used to report or release the rectification plan data of the system functions to be improved according to the data maintenance reporting rules; The historical data maintenance form processing module is specifically used to traverse and match the historical data maintenance forms of the target time window based on the natural language processing technology to determine similar historical data maintenance forms; classify the similar historical data maintenance forms according to the large language model to determine the historical data to be matched; The similarity matching module includes a first matching unit and a second matching unit; the first matching unit is used to perform content similarity matching between the target data maintenance form currently submitted by the user and the historical data to be matched, so as to obtain SQL data to be matched; The second matching unit is used to perform SQL similarity matching on the SQL data to be matched, determine a repeated triggering data maintenance sheet, and determine the system function to be improved according to the repeated triggering data maintenance sheet.

7. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can perform the data maintenance method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the data maintenance method according to any one of claims 1 to 5 when executed.

Images