File comparison method, apparatus, device, medium, and product

Abstract

The application belongs to the technical field of artificial intelligence, and specifically provides a file comparison method, device, equipment, medium and product. The method comprises the following steps: obtaining a transaction verification file and a transaction single file to be processed; performing semantic recognition and element structuring processing on the transaction verification file and the transaction single file by using a preset semantic recognition and extraction model, to generate first structured element data corresponding to the transaction verification file and second structured element data corresponding to the transaction single file; the first structured element data comprises first key element data; the second structured element data comprises second key element data; determining matched first structured element data and second structured element data according to the first key element data and the second key element data; comparing the matched first structured element data and the second structured element data to generate a corresponding comparison result. The file comparison method of the application is shorter in time and higher in processing efficiency compared with the manual method.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of artificial intelligence, and in particular to a file comparison method and device, equipment, medium and product. BACKGROUND

[0002] In the financial field, such as banks, business personnel need to take the transaction verification file provided by the transaction party offline to determine whether a transaction is correct and consistent. According to some key elements in the transaction verification file, the current verification transaction file is found in the system by the naked eye, and other information of the transaction verification file and the transaction file is compared to ensure the correctness and consistency of the transaction data.

[0003] At present, the comparison between the transaction verification file and the transaction file is carried out by manual method, which takes about 7 minutes on average each time, and the processing efficiency is low. SUMMARY

[0004] The present application provides a file comparison method, device, equipment, medium and product to solve the problem of long time consumption and low processing efficiency in the comparison between the transaction verification file and the transaction file by manual method.

[0005] The first aspect of the present application provides a file comparison method, comprising:

[0006] Obtaining a transaction verification file and a transaction file to be processed;

[0007] Using a preset semantic recognition and extraction model to perform semantic recognition and element structuring processing on the transaction verification file and the transaction file, to generate first structured element data corresponding to the transaction verification file and second structured element data corresponding to the transaction file; the first structured element data includes first key element data; the second structured element data includes second key element data;

[0008] According to the first key element data and the second key element data, the first structured element data and the second structured element data are determined to be matched;

[0009] Comparing the matched first structured element data and the second structured element data to generate a corresponding comparison result.

[0010] Further, the method as described above, the obtaining of the transaction verification file and the transaction file to be processed, comprises:

[0011] Receiving a transaction verification file to be processed uploaded by a robot process automation application;

[0012] Obtaining a transaction file to be processed from a front-end transaction server.

[0013] Further, the method as described above, the preset semantic recognition and extraction model comprises: an NLP semantic recognition model and an element analysis and extraction model;

[0014] The semantic recognition and element structuring processing of the transaction certification file and the transaction single file by using the preset semantic recognition and extraction model comprises:

[0015] The semantic recognition and element extraction of the transaction certification file and the transaction single file by using the NLP semantic recognition model comprises:

[0016] The data cleaning and element analysis processing of the first element text data and the second element text data by using the element analysis and extraction model comprises:

[0017] Further, the method as described above, the semantic recognition and element extraction of the transaction certification file and the transaction single file by using the NLP semantic recognition model comprises:

[0018] The semantic recognition of the transaction certification file and the transaction single file by using the NLP semantic recognition model comprises:

[0019] The element extraction of each element in the first total text data and the second total text data by using the NLP semantic recognition model comprises:

[0020] Further, the method as described above, the data cleaning and element analysis processing of the first element text data and the second element text data by using the element analysis and extraction model comprises:

[0021] The data cleaning of the first element text data and the second element text data by using the element analysis and extraction model comprises:

[0022] The element analysis of the cleaned first element text data and the cleaned second element text data by using the element analysis and extraction model comprises:

[0023] Further, the method as described above, the first structured element data further comprises: first common element data; and the second structured element data further comprises: second common element data.

[0024] The element analysis and extraction model is used to perform element analysis on the cleaned first element text data and the cleaned second element text data, to generate the first structured element data and the second structured element data, including:

[0025] The element analysis and extraction model determines the first key element data and the first common element data in the cleaned first element text data based on the first business category; the first business category is the business category of the transaction verification file corresponding to the first element text data.

[0026] The element analysis and extraction model determines the second key element data and the second common element data in the cleaned second element text data based on the second business category; the second business category is the business category of the transaction single file corresponding to the second element text data.

[0027] Further, the method as described above, the first structured element data further comprises: first common element data; and the second structured element data further comprises: second common element data.

[0028] The first common element data and the second common element data are compared and matched.

[0029] If it is determined that the matched first common element data and the second common element data are inconsistent, it is determined whether there is a corresponding fuzzy matching strategy in the preset database according to the first business category and the second business category.

[0030] If it is determined that there is a fuzzy matching strategy, the matched first common element data and the second common element data are fuzzy matched according to the fuzzy matching strategy, to generate a fuzzy matching comparison result.

[0031] If it is determined that the matched first common element data and the second common element data are consistent, a consistent comparison result is generated.

[0032] Further, the method as described above, if the fuzzy matching comparison result is fuzzy matching inconsistency, the method further comprises:

[0033] The first common element data and the second common element data of the fuzzy matching inconsistency are output.

[0034] The second aspect of the present application provides a file comparison device, comprising:

[0035] An acquisition module is configured to acquire a transaction verification file and a transaction single file to be processed.

[0036] generating a first structured element data corresponding to the transaction certification file and a second structured element data corresponding to the transaction order file by performing semantic recognition and element structuring on the transaction certification file and the transaction order file using a preset semantic recognition extraction model; the first structured element data comprises first key element data; and the second structured element data comprises second key element data;

[0037] determining matched first structured element data and second structured element data according to the first key element data and the second key element data;

[0038] comparing the matched first structured element data and the second structured element data to generate a corresponding comparison result.

[0039] Further, the apparatus described above, the acquisition module is specifically used for:

[0040] receiving a transaction certification file to be processed uploaded by a robot process automation application; and acquiring a transaction order file to be processed from a foreground transaction server.

[0041] Further, the apparatus described above, the preset semantic recognition extraction model comprises an NLP semantic recognition model and an element analysis extraction model.

[0042] The generation module is specifically used for:

[0043] performing semantic recognition and element extraction on the transaction certification file and the transaction order file using the NLP semantic recognition model to generate first element text data corresponding to the transaction certification file and second element text data corresponding to the transaction order file; and performing data cleaning and element analysis on the first element text data and the second element text data using the element analysis extraction model to generate the first structured element data and the second structured element data.

[0044] Further, the apparatus described above, when the generation module performs semantic recognition and element extraction on the transaction certification file and the transaction order file using the NLP semantic recognition model to generate first element text data corresponding to the transaction certification file and second element text data corresponding to the transaction order file, the generation module is specifically used for:

[0045] performing semantic recognition on the transaction certification file and the transaction order file using the NLP semantic recognition model to generate first total text data corresponding to the transaction certification file and second total text data corresponding to the transaction order file; and performing element extraction on each element in the first total text data and the second total text data using the NLP semantic recognition model to generate the first element text data and the second element text data.

[0046] Further, the apparatus as described above, the generation module is specifically used for:

[0047] The element analysis extraction model is used to clean up the first element text data and the second element text data, and generate cleaned first element text data and cleaned second element text data; the element analysis extraction model is used to analyze the cleaned first element text data and the cleaned second element text data, and generate the first structured element data and the second structured element data.

[0048] Further, the apparatus as described above, the first structured element data further comprises: first general element data; and the second structured element data further comprises: second general element data.

[0049] The generation module is specifically used for:

[0050] The element analysis extraction model is used to determine the first key element data and the first general element data in the cleaned first element text data based on the first business category; the first business category is the business category of the transaction verification file corresponding to the first element text data; the element analysis extraction model is used to determine the second key element data and the second general element data in the cleaned second element text data based on the second business category; the second business category is the business category of the transaction single file corresponding to the second element text data.

[0051] Further, the apparatus as described above, the comparison module is specifically used for:

[0052] The first general element data and the second general element data are compared and matched; if it is determined that the matched first general element data and the second general element data are inconsistent, it is determined whether there is a corresponding fuzzy matching strategy in the preset database according to the first business category and the second business category; if it is determined that there is a fuzzy matching strategy, the matched first general element data and the second general element data are fuzzy matched according to the fuzzy matching strategy, and a fuzzy matched comparison result is generated; if it is determined that the matched first general element data and the second general element data are consistent, a consistent comparison result is generated.

[0053] Further, the apparatus as described above, if the fuzzy matched comparison result is fuzzy matching inconsistency, the comparison module is further used for:

[0054] outputting the first common element data and the second common element data that are inconsistent in fuzzy matching.

[0055] The third aspect of the present application provides an electronic device, comprising a memory and a processor.

[0056] The memory stores computer execution instructions.

[0057] The processor executes the computer execution instructions stored in the memory to implement the file comparison method according to any one of the first aspect.

[0058] The fourth aspect of the present application provides a computer readable storage medium, wherein the computer readable storage medium stores computer execution instructions, and the computer execution instructions are executed by a processor to implement the file comparison method according to any one of the first aspect.

[0059] The fifth aspect of the present application provides a computer program product, comprising a computer program, wherein the computer program is executed by a processor to implement the file comparison method according to any one of the first aspect.

[0060] The file comparison method, device, equipment, medium and product provided by the present application, the method comprises: obtaining a transaction confirmation file and a transaction single file to be processed; using a preset semantic recognition extraction model to perform semantic recognition and element structuring processing on the transaction confirmation file and the transaction single file, to generate first structured element data corresponding to the transaction confirmation file and second structured element data corresponding to the transaction single file; the first structured element data comprises first key element data; the second structured element data comprises second key element data; determining matched first structured element data and second structured element data according to the first key element data and the second key element data; comparing the matched first structured element data and the second structured element data to generate a corresponding comparison result. The file comparison method of the present application uses a preset semantic recognition extraction model to perform semantic recognition and element structuring processing on the transaction confirmation file and the transaction single file, to generate first structured element data corresponding to the transaction confirmation file and second structured element data corresponding to the transaction single file. Thus, based on the first structured element data and the second structured element data, a corresponding comparison result is generated. Compared with the manual method, the time consumption is shorter and the processing efficiency is higher. BRIEF DESCRIPTION OF DRAWINGS

[0061] The accompanying drawings, which are incorporated into and form part of the specification, illustrate embodiments consistent with the present application and, together with the specification, serve to explain the principles of the present application.

[0062] Figure 1 A scene diagram of the file comparison method according to the embodiments of the present application is shown.

[0063] Figure 2 Flowchart of the file comparison method provided in the present application Figure 1 ;

[0064] Figure 3 Flowchart of the file comparison method provided in the present application Figure 2 ;

[0065] Figure 4 Overall flowchart of the file comparison method provided in the present application

[0066] Figure 5 Structure diagram of the file comparison device provided in the present application

[0067] Figure 6 Structure diagram of the electronic device provided in the present application.

[0068] The specific embodiments of the present application have been shown in the above-described drawings, and will be described in more detail hereinafter. These drawings and the written description are not intended to restrict the scope of the present application concept in any way, but to illustrate the present application concept to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0069] The exemplary embodiments will be described in detail herein with reference to the attached drawings. In the following description, the same numbers are used to indicate the same or similar components. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with the present application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of the present application as detailed in the appended claims.

[0070] In the technical solutions of the embodiments of the present application, the collection, storage, use, processing, transmission, provision and disclosure of user personal information comply with relevant laws and regulations and do not violate public order and good customs.

[0071] It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of relevant data need to comply with relevant laws and regulations and standards of relevant countries and regions, and provide corresponding operation portal for user to choose authorization or refusal.

[0072] It should be noted that the comparison method, device, equipment, medium and product of the present disclosure can be used in the field of artificial intelligence technology. It can also be used in any field other than the field of artificial intelligence technology. The application field of the comparison method, device, equipment, medium and product of the present disclosure is not limited.

[0073] The technical solutions of the present application will be described in detail below with specific examples. The following specific examples can be combined with each other, and the same or similar concepts or processes may not be described again in some examples. The embodiments of the present application will be described below with reference to the accompanying drawings.

[0074] In order to clearly understand the technical solutions of the present application, the prior art solutions will be described in detail first. At present, the comparison between the transaction confirmation file and the transaction single file is carried out by manual method. The business personnel not only need to determine the matching transaction single file of the transaction confirmation file by naked eye, but also need to further compare the consistency between the matching transaction single files of the transaction confirmation file. Because there are many elements in the transaction confirmation file and the transaction single file, the business personnel usually need to spend a long time, for example, the average comparison time of each transaction is about 7 minutes, the average daily transaction is about 20, and the business personnel need to spend 140 minutes per day, which is time-consuming and inefficient.

[0075] Therefore, in order to solve the problem of long time consumption and low processing efficiency in the prior art comparison between the transaction confirmation file and the transaction single file by manual method, the inventors found in the research that in order to solve the problem, an automatic processing method can be used to automatically obtain the transaction confirmation file and the transaction single file to be processed, automatically compare the transaction confirmation file and the transaction single file to reduce the time consumption and improve the processing efficiency.

[0076] Specifically, the transaction confirmation file and the transaction single file to be processed are obtained. A preset semantic recognition and extraction model is used to perform semantic recognition and element structuring processing on the transaction confirmation file and the transaction single file, to generate first structured element data corresponding to the transaction confirmation file and second structured element data corresponding to the transaction single file. The first structured element data includes first key element data. The second structured element data includes second key element data. The matching first structured element data and second structured element data are determined according to the first key element data and the second key element data. The matching first structured element data and second structured element data are compared to generate a corresponding comparison result.

[0077] The file comparison method of the present application adopts a preset semantic recognition extraction model to perform semantic recognition and element structuring processing on the transaction verification file and the transaction order file, to generate first structured element data corresponding to the transaction verification file and second structured element data corresponding to the transaction order file. Then, the first structured element data and the second structured element data are compared to generate a corresponding comparison result. Compared with the manual method, the present application has shorter time consumption and higher processing efficiency.

[0078] Based on the above creative findings, the inventors proposed the technical solution of the present application.

[0079] The application scenarios of the file comparison method provided by the embodiments of the present application will be introduced below. As shown in Figure 1 The network architecture of the application scenario corresponding to the file comparison method provided by the embodiments of the present application includes the first electronic device 1 and the second electronic device 2. The second electronic device 2 stores the transaction verification file and the transaction order file to be processed, which can be obtained from the front-end transaction server.

[0080] For example, the second electronic device 2 sends the transaction verification file and the transaction order file to be processed to the first electronic device 1. The first electronic device 1 adopts a preset semantic recognition extraction model to perform semantic recognition and element structuring processing on the transaction verification file and the transaction order file, to generate first structured element data corresponding to the transaction verification file and second structured element data corresponding to the transaction order file. The first structured element data includes first key element data. The second structured element data includes second key element data. At the same time, the first electronic device 1 determines the matching first structured element data and second structured element data according to the first key element data and the second key element data. The first electronic device 1 compares the matching first structured element data and second structured element data to generate a corresponding comparison result. The comparison result can be consistent or inconsistent between the first structured element data and the second structured element data, which represents that the transaction verification file corresponding to the first structured element data and the transaction order file corresponding to the second structured element data are consistent or inconsistent. After determining the comparison result, the comparison result can be displayed to the business personnel, or output to the business personnel terminal device to enable the business personnel to know the comparison result and perform subsequent processing.

[0081] The embodiments of the present application will be introduced below in conjunction with the drawings of the specification.

[0082] Figure 2 The flowchart of the file comparison method provided by the present application Figure 1 As shown in Figure 2As shown, in this embodiment, the execution subject of the embodiment of the application is a file comparison device, which can be integrated in an electronic device. The file comparison method provided in this embodiment includes the following steps:

[0083] In step S101, a transaction verification file to be processed and a transaction single file are obtained.

[0084] In this embodiment, the transaction verification file to be processed can be obtained through a robot process automation application or directly received from a transaction device.

[0085] The transaction single file can be obtained from a front-end transaction server.

[0086] The transaction verification file and the transaction single file can be one or more.

[0087] In step S102, a preset semantic recognition and extraction model is used to perform semantic recognition and element structuring processing on the transaction verification file and the transaction single file, to generate first structured element data corresponding to the transaction verification file and second structured element data corresponding to the transaction single file. The first structured element data includes first key element data. The second structured element data includes second key element data.

[0088] In this embodiment, the preset semantic recognition and extraction model can include a semantic recognition model and an element analysis and extraction model. The semantic recognition model is used to perform semantic recognition on the transaction verification file and the transaction single file. The element analysis and extraction model is used to extract elements and realize element structuring.

[0089] The preset semantic recognition and extraction model can be trained according to historical transaction verification files and historical transaction single files, thereby improving the accuracy of recognition and element extraction.

[0090] The transaction verification file and the transaction single file have various elements, some of which are key elements, i.e., the first key element data and the second key element data mentioned above. The first key element data is, for example, the time in the transaction verification file, the identity of the transaction parties, etc. The second key element data is, for example, the transaction amount, the transaction time, the identification of the transaction parties, etc. According to different business categories, the key element data can change, and can be set according to actual application.

[0091] In step S103, the first structured element data and the second structured element data that match are determined according to the first key element data and the second key element data.

[0092] In this embodiment, the first key element data is matched with the corresponding second key element data, so as to determine the matched first key element data and the second key element data. According to the matched first key element data and the second key element data, the matched first structured element data and the second structured element data can be further determined.

[0093] In step S104, the matched first structured element data and the second structured element data are compared to generate a corresponding comparison result.

[0094] The comparison of the matched first structured element data and the second structured element data can be consistent or inconsistent. It represents that the corresponding transaction verification file and the corresponding transaction single file are consistent or inconsistent.

[0095] The file comparison method provided by the embodiment of the application comprises the following steps.

[0096] The file comparison method provided by the embodiment of the application comprises the following steps.

[0097] Figure 3 The file comparison method provided by the embodiment of the application comprises the following steps. Figure 2 As shown in Figure 3 The file comparison method provided by the embodiment of the application is further refined on the basis of the file comparison method provided by the previous embodiment of the application. The file comparison method provided by the embodiment of the application comprises the following steps.

[0098] In step S201, the transaction verification file to be processed uploaded by the robot process automation application is received.

[0099] In this embodiment, the request for obtaining the transaction verification file does not need to be sent, and the corresponding transaction verification file can be uploaded through the robotic process automation application, so that the efficiency of obtaining the transaction verification file can be further improved.

[0100] In step S202, the transaction sheet file to be processed is obtained from the foreground transaction server.

[0101] In this embodiment, the foreground transaction server receives various transaction sheet files, and the transaction sheet file to be processed can be directly obtained from the foreground transaction server each time a transaction sheet file is received, so that the efficiency of obtaining the transaction sheet file can be improved.

[0102] It should be noted that the preset semantic recognition and extraction model includes an NLP semantic recognition model and an element analysis and extraction model.

[0103] In step S203, the NLP semantic recognition model is used to perform semantic recognition and element extraction on the transaction verification file and the transaction sheet file, to generate first element text data corresponding to the transaction verification file and second element text data corresponding to the transaction sheet file.

[0104] In this embodiment, NLP (English full name: Natural Language Processing, Chinese: Natural Language Processing) can perform semantic recognition and element extraction processing on the transaction verification file and the transaction sheet file after the NLP semantic recognition model is pre-trained. The accuracy of semantic recognition and element extraction processing is relatively high.

[0105] The first element text data corresponding to the transaction verification file is complete element data of the transaction verification file, and the second element text data corresponding to the transaction sheet file is complete element data of the transaction sheet file. Since the transaction verification file and the transaction sheet file not only include elements but also include frames, layouts, and texts, complete element data can be extracted through the NLP semantic recognition model, which facilitates subsequent element comparison.

[0106] Optionally, in this embodiment, step S203 can be specifically as follows:

[0107] The NLP semantic recognition model is used to perform semantic recognition on the transaction verification file and the transaction sheet file, to generate first total text data corresponding to the transaction verification file and second total text data corresponding to the transaction sheet file.

[0108] The NLP semantic recognition model is used to perform element extraction on each element in the first total text data and the second total text data, to generate the first element text data and the second element text data.

[0109] According to the actual application scene, the element data in the transaction verification file and the transaction single file is labeled in advance, including the specific element name and the expression style of the element. The NLP semantic recognition model is established, the element extraction rule is defined, and the transaction verification file and the transaction single file are trained. At the same time, according to the nature of the element, the matching type and model are selected, the related elements are extracted, the structured data is generated, and the specific content of the extraction is adjusted according to the labeled transaction verification file and the transaction single file to optimize the rule and the model.

[0110] In step S204, the element analysis extraction model is used to perform data cleaning and element analysis processing on the first element text data and the second element text data to generate first structured element data and second structured element data.

[0111] In this embodiment, data cleaning refers to finding and correcting identifiable errors in data files, including checking data consistency, handling invalid values and missing values, etc. Through data cleaning processing, the error data in the first element text data and the second element text data can be removed, providing a basis for subsequent comparison and avoiding interference caused by error data on comparison, thereby improving the accuracy of subsequent comparison. Element analysis processing mainly classifies and structures the elements, which facilitates subsequent comparison processing and improves comparison efficiency.

[0112] Optionally, in this embodiment, step S204 can be specifically:

[0113] The element analysis extraction model is used to perform data cleaning on the first element text data and the second element text data to generate cleaned first element text data and cleaned second element text data.

[0114] The element analysis extraction model is used to perform element analysis on the cleaned first element text data and the cleaned second element text data to generate first structured element data and second structured element data.

[0115] The cleaned first element text data and the cleaned second element text data remove error data, and the element analysis based on the cleaned first element text data and the cleaned second element text data can determine more accurate first structured element data and second structured element data.

[0116] Optionally, in this embodiment, the first structured element data further includes first normal element data. The second structured element data further includes second normal element data.

[0117] At the same time, the process of using the element analysis extraction model to perform element analysis on the cleaned first element text data and the cleaned second element text data to generate first structured element data and second structured element data can be specifically:

[0118] The element analysis extraction model determines first key element data and first normal element data in the cleaned first element text data based on the first business category. The first business category is a business category of the transaction verification file corresponding to the first element text data.

[0119] The element analysis extraction model determines second key element data and second normal element data in the cleaned second element text data based on the second business category. The second business category is a business category of the transaction single file corresponding to the second element text data.

[0120] The first normal element data is element data in the first structured element data except the first key element data, which is less important than the first key element data. Similarly, the second normal element data is element data in the second structured element data except the second key element data.

[0121] In this embodiment, the key element data is related to the business category, and different business categories can set different key elements. For example, for a transfer transaction, the key element data can be the account of the transfer parties and the transfer row number, and for a financial transaction, it can be the transaction item and the input amount.

[0122] When performing element analysis, the parsed element data can be structured by a preset field configuration function to form unified structured data for subsequent element comparison.

[0123] In step S205, the first structured element data and the second structured element data that match are determined according to the first key element data and the second key element data.

[0124] In this embodiment, the implementation manner of step 205 is similar to that of step 103 of the previous embodiment, which will not be repeated here.

[0125] In step S206, the matched first structured element data and second structured element data are compared to generate a corresponding comparison result.

[0126] In this embodiment, the implementation manner of step 206 is similar to that of step 104 of the previous embodiment, which will not be repeated here.

[0127] Optionally, in this embodiment, step S206 can be specifically:

[0128] The matched first normal element data and second normal element data are compared.

[0129] If it is determined that the matched first common element data and the second common element data are inconsistent, it is determined whether there is a corresponding fuzzy matching strategy in the preset database according to the first business category and the second business category.

[0130] If it is determined that there is a fuzzy matching strategy, the matched first common element data and the second common element data are subjected to fuzzy matching according to the fuzzy matching strategy, and a fuzzy matching comparison result is generated.

[0131] If it is determined that the matched first common element data and the second common element data are consistent, a consistent comparison result is generated.

[0132] Optionally, in the embodiment, if the fuzzy matching comparison result is inconsistent fuzzy matching, the inconsistent fuzzy matching first common element data and the second common element data are output.

[0133] Some business categories have higher precision requirements and cannot use the fuzzy matching method, and some business categories have relatively low precision requirements and can use the fuzzy matching method. When comparing the first common element data and the second common element data, if the comparison result is inconsistent, it is checked whether there is a fuzzy matching strategy, if there is, the fuzzy matching is continued, if there is not, it is set to manual comparison. If the fuzzy matching is inconsistent, the inconsistent fuzzy matching first common element data and the second common element data are output, so as to perform manual comparison, if they are consistent, it is determined that the automatic matching is successful. At the same time, the element results of the two parties in the comparison can also be displayed on the display page, the inconsistent results are marked yellow, and the business personnel are prompted.

[0134] The overall flow of the file comparison method of the embodiment is as shown in Figure 4 The overall flow of the embodiment is as follows:

[0135] The transaction confirmation file and the transaction single file are obtained, NLP semantic recognition is performed on the transaction confirmation file and the transaction single file, and transaction confirmation file recognition results and transaction single file recognition results are generated. At the same time, element analysis and extraction are performed on the transaction confirmation file recognition results and the transaction single file recognition results, and the extracted results are subjected to comparison processing. If the comparison result is consistent, it is determined that the automatic comparison processing is completed, if the comparison result is inconsistent, it is converted into manual processing, so as to improve the accuracy of the comparison.

[0136] The method of the embodiment completes the comparison of the online transaction confirmation through automatic comparison of the transaction confirmation file and the transaction single file, reduces the manual cost of the business, improves the maintainability of the system, and at the same time reduces the probability of manual visual checking errors, and improves the processing efficiency.

[0137] Figure 5 The structure diagram of the file comparison device provided in the application is as shown in Figure 5As shown, in this embodiment, the file comparison device 300 can be arranged in an electronic device, and the file comparison device 300 comprises:

[0138] The acquisition module 301 is configured to acquire a transaction confirmation file and a transaction slip file to be processed.

[0139] The generation module 302 is configured to perform semantic recognition and element structuring processing on the transaction confirmation file and the transaction slip file by using a preset semantic recognition extraction model, to generate first structured element data corresponding to the transaction confirmation file and second structured element data corresponding to the transaction slip file. The first structured element data comprises first key element data. The second structured element data comprises second key element data.

[0140] The determination module 303 is configured to determine matched first structured element data and second structured element data according to the first key element data and the second key element data.

[0141] The comparison module 304 is configured to compare the matched first structured element data and the second structured element data, to generate a corresponding comparison result.

[0142] The file comparison device provided in this embodiment can perform Figure 2 The technical solutions of the method embodiment are similar to those of the file comparison device embodiment, and thus the implementation principles and technical effects are similar to those of the file comparison device embodiment, which will not be repeated here. Figure 2

[0143] The file comparison device provided in this embodiment is further refined on the basis of the file comparison device provided in the previous embodiment, and thus the file comparison device 300 comprises:

[0144] Optionally, in this embodiment, the acquisition module 301 is specifically configured to:

[0145] The robot process automation application uploads a transaction confirmation file to be processed. The transaction slip file to be processed is acquired from a front-end transaction server.

[0146] Optionally, in this embodiment, the preset semantic recognition extraction model comprises an NLP semantic recognition model and an element parsing extraction model.

[0147] The generation module 302 is specifically configured to:

[0148] The NLP semantic recognition model is used to perform semantic recognition and element extraction on the transaction confirmation file and the transaction slip file, to generate first element text data corresponding to the transaction confirmation file and second element text data corresponding to the transaction slip file. The element parsing extraction model is used to perform data cleaning and element parsing processing on the first element text data and the second element text data, to generate the first structured element data and the second structured element data.​

[0149] Optionally, in the embodiment, the generation module 302, when generating the first element text data corresponding to the transaction certification document and the second element text data corresponding to the transaction document by using the NLP semantic recognition model to perform semantic recognition and element extraction on the transaction certification document and the transaction document, is specifically configured to:

[0150] perform semantic recognition on the transaction certification document and the transaction document by using the NLP semantic recognition model to generate the first total text data corresponding to the transaction certification document and the second total text data corresponding to the transaction document, and perform element extraction on each element in the first total text data and the second total text data by using the NLP semantic recognition model to generate the first element text data and the second element text data.

[0151] Optionally, in the embodiment, the generation module 302, when generating the first structured element data and the second structured element data by using the element analysis extraction model to perform data cleaning and element analysis processing on the first element text data and the second element text data, is specifically configured to:

[0152] perform data cleaning on the first element text data and the second element text data by using the element analysis extraction model to generate the cleaned first element text data and the cleaned second element text data, and perform element analysis on the cleaned first element text data and the cleaned second element text data by using the element analysis extraction model to generate the first structured element data and the second structured element data.

[0153] Optionally, in the embodiment, the first structured element data further includes first common element data, and the second structured element data further includes second common element data.

[0154] The generation module 302, when generating the first structured element data and the second structured element data by using the element analysis extraction model to perform element analysis on the cleaned first element text data and the cleaned second element text data, is specifically configured to:

[0155] determine, based on the first business category, the first key element data and the first common element data in the cleaned first element text data by using the element analysis extraction model. The first business category is a business category of the transaction certification document corresponding to the first element text data. Determine, based on the second business category, the second key element data and the second common element data in the cleaned second element text data by using the element analysis extraction model. The second business category is a business category of the transaction document corresponding to the second element text data.

[0156] Optionally, in the embodiment, the comparison module 304 is specifically configured to:

[0157] The system compares and matches the first and second ordinary feature data. If the first and second ordinary feature data are found to be inconsistent, it determines whether a corresponding fuzzy matching strategy exists in the preset database based on the first and second business categories. If a fuzzy matching strategy exists, it performs a fuzzy match on the first and second ordinary feature data according to the strategy, generating a fuzzy matching comparison result. If the first and second ordinary feature data are found to be consistent, a consistent comparison result is generated.

[0158] Optionally, in this embodiment, if the comparison result of the fuzzy matching is inconsistent, the comparison module 304 is further used to:

[0159] Output the first and second ordinary feature data that do not match the fuzzy match.

[0160] The document comparison device provided in this embodiment can perform... Figures 2-4 The technical solution of the method embodiment shown has the same implementation principle and technical effect as... Figures 2-4 The methods and embodiments shown are similar and will not be described in detail here.

[0161] According to embodiments of this application, this application also provides an electronic device, a computer-readable storage medium, and a computer program product.

[0162] like Figure 6 As shown, Figure 6 This is a schematic diagram of the electronic device provided in this application. The electronic device is intended for various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, blade servers, mainframe computers, and other suitable computers. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present application described and / or claimed herein.

[0163] like Figure 6 As shown, the electronic device includes a processor 401 and a memory 402. The various components are interconnected via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processor can process instructions executed within the electronic device.

[0164] The memory 402 is the non-transitory computer-readable storage medium provided in this application. The memory stores instructions executable by at least one processor to cause at least one processor to perform the file comparison method provided in this application. The non-transitory computer-readable storage medium of this application stores computer instructions for causing a computer to perform the file comparison method provided in this application.

[0165] Memory 402, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as the program instructions / modules corresponding to the file comparison method in the embodiments of this application (e.g., appendix). Figure 5 The acquisition module 301, generation module 302, determination module 303, and comparison module 304 are shown. The processor 401 executes various functional applications and data processing of the electronic device by running non-transient software programs, instructions, and modules stored in the memory 402, thereby implementing the file comparison method in the above method embodiment.

[0166] In addition, this embodiment also provides a computer product, which, when the instructions in the computer product are executed by the processor of an electronic device, enables the electronic device to execute the file comparison method of the above embodiment.

[0167] Other embodiments of the present application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the embodiments of this application that follow the general principles of the embodiments of this application and include common knowledge or customary techniques in the art not disclosed in the embodiments of this application.

[0168] It should be understood that the embodiments of this application are not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from their scope. The scope of the embodiments of this application is limited only by the appended claims.

Claims

1. A method of file comparison, characterized by, The method comprises the following steps: obtaining a transaction confirmation file and a transaction order file to be processed; performing semantic recognition and element structuring processing on the transaction confirmation file and the transaction order file by using a preset semantic recognition and extraction model to generate first structured element data corresponding to the transaction confirmation file and second structured element data corresponding to the transaction order file; the first structured element data comprises first key element data; the second structured element data comprises second key element data; determining matched first structured element data and second structured element data according to the first key element data and the second key element data; comparing the matched first structured element data and the second structured element data to generate a corresponding comparison result; the first structured element data further comprises first general element data; the second structured element data further comprises second general element data; comparing the matched first general element data and the second general element data to generate a corresponding comparison result, comprising: comparing the matched first general element data and the second general element data; if it is determined that the matched first general element data and the second general element data are inconsistent, determining whether there is a corresponding fuzzy matching strategy in a preset database according to a first business category and a second business category, the first business category being a business category of the transaction confirmation file corresponding to the first element text data, the second business category being a business category of the transaction order file corresponding to the second element text data, the first element text data and the second element text data being obtained by performing semantic recognition and element extraction on the transaction confirmation file and the transaction order file by using an NLP semantic recognition model in the preset semantic recognition and extraction model; if it is determined that there is a fuzzy matching strategy, performing fuzzy matching on the matched first general element data and the second general element data according to the fuzzy matching strategy to generate a fuzzy matching comparison result; if it is determined that the matched first general element data and the second general element data are consistent, generating a consistent comparison result.

2. The method of claim 1, wherein, The method comprises the following steps: receiving a transaction confirmation file to be processed uploaded by a robot process automation application; obtaining a transaction order file to be processed from a front-end transaction server.

3. The method according to claim 1 or 2, characterized in that, The preset semantic recognition and extraction model comprises an NLP semantic recognition model and an element analysis and extraction model; The method comprises the following steps: performing semantic recognition and element extraction on the transaction confirmation file and the transaction order file by using the NLP semantic recognition model to generate first element text data corresponding to the transaction confirmation file and second element text data corresponding to the transaction order file; performing data cleaning and element analysis processing on the first element text data and the second element text data by using the element analysis and extraction model to generate the first structured element data and the second structured element data.

4. The method of claim 3, wherein, The semantic recognition and element extraction of the transaction certification file and the transaction single file by using the NLP semantic recognition model generate the first element text data corresponding to the transaction certification file and the second element text data corresponding to the transaction single file, which comprises: The semantic recognition of the transaction certification file and the transaction single file by using the NLP semantic recognition model generates the first total text data corresponding to the transaction certification file and the second total text data corresponding to the transaction single file; The element extraction of each element in the first total text data and the second total text data by using the NLP semantic recognition model generates the first element text data and the second element text data.

5. The method of claim 3, wherein, The data cleaning and element analysis processing of the first element text data and the second element text data by using the element analysis extraction model generate the first structured element data and the second structured element data, which comprises: The data cleaning of the first element text data and the second element text data by using the element analysis extraction model generates the cleaned first element text data and the cleaned second element text data; The element analysis of the cleaned first element text data and the cleaned second element text data by using the element analysis extraction model generates the first structured element data and the second structured element data.

6. The method of claim 5, wherein, The element analysis of the cleaned first element text data and the cleaned second element text data by using the element analysis extraction model generates the first structured element data and the second structured element data, which comprises: The first key element data and the first normal element data in the cleaned first element text data are determined based on the first business category by using the element analysis extraction model, and the second key element data and the second normal element data in the cleaned second element text data are determined based on the second business category by using the element analysis extraction model.

7. The method of claim 1, wherein, If the comparison result of the fuzzy matching is inconsistent, the method further comprises: Output the first normal element data and the second normal element data which are inconsistent in fuzzy matching.

8. A file comparison apparatus characterized by comprising: Comprise: An acquisition module is configured to acquire a transaction certification file and a transaction single file to be processed; A generation module is configured to perform semantic recognition and element structuring processing on the transaction certification file and the transaction single file by using a preset semantic recognition extraction model, and generate first structured element data corresponding to the transaction certification file and second structured element data corresponding to the transaction single file; the first structured element data comprises first key element data; and the second structured element data comprises second key element data; A determination module is configured to determine matched first structured element data and second structured element data according to the first key element data and the second key element data; A comparison module is configured to compare the matched first structured element data and the second structured element data to generate a corresponding comparison result; The first structured element data further comprises first normal element data; and the second structured element data further comprises second normal element data; The comparison module is specifically configured to: Compare the matched first normal element data and the second normal element data; If it is determined that the matched first common element data and the second common element data are inconsistent, it is determined whether there is a corresponding fuzzy matching strategy in the preset database according to the first business category and the second business category, the first business category being a business category of the transaction verification file corresponding to the first element text data, the second business category being a business category of the transaction sheet file corresponding to the second element text data, the first element text data and the second element text data being obtained by performing semantic recognition and element extraction on the transaction verification file and the transaction sheet file by using an NLP semantic recognition model in the preset semantic recognition and extraction model; If it is determined that there is a fuzzy matching strategy, the matched first common element data and the second common element data are matched according to the fuzzy matching strategy, and a fuzzy matching comparison result is generated. If it is determined that the matched first common element data and the second common element data are consistent, a consistent comparison result is generated.

9. An electronic device, comprising: Comprise: a memory and a processor; the memory stores computer execution instructions; the processor executes the computer execution instructions stored in the memory to realize the file comparison method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer execution instructions, and the computer execution instructions are executed by the processor to realize the file comparison method according to any one of claims 1 to 7.

11. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to realize the file comparison method according to any one of claims 1 to 7.

Images