Field mapping method, device and equipment and computer readable storage medium

Abstract

The invention discloses a field mapping method, device and equipment and a computer readable storage medium, which are applied to the technical field of computers, and comprise the following steps: obtaining a to-be-matched field; matching the to-be-matched fields by utilizing a pre-training semantic model and a semantic rule library to obtain a matching result; and determining corresponding fields according to the matching result, and giving a correction suggestion when a conflict problem exists between the corresponding fields to obtain a corrected matching result. According to the method, comprehensive field matching is carried out based on the pre-trained semantic model and the semantic rule base, conflict problem detection and intelligent correction are carried out on the fields, the field isomerism problem in cross-channel data integration is solved, zero-code rapid mapping is achieved, and the data standardization efficiency and accuracy are improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and particularly to a field mapping method, apparatus, device and computer-readable storage medium. Background Art

[0002] With the development of Internet technology and big data, the advertising industry has achieved multi-platform and multi-channel development. Correspondingly, the channels of advertising placement data are becoming more and more extensive. In the advertising placement scenario, there are problems such as differences in field naming, inconsistent statistical dimensions, and data format conflicts in the data sources of multiple advertising channels. These problems lead to field heterogeneity in cross-channel data integration, low efficiency and poor accuracy of data standardization.

[0003] Traditional field mapping methods mainly rely on manual operations, which have problems such as low efficiency, poor scalability, high cost, and difficult to guarantee data quality. Therefore, there is a need for an efficient, scalable, low-cost field mapping method that can guarantee data quality to solve the field heterogeneity problem in cross-channel data integration and improve the efficiency and accuracy of data standardization. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide a field mapping method, apparatus, device and computer-readable storage medium, which solves the problem of low efficiency of field mapping in the prior art.

[0005] To solve the above technical problems, the present invention provides a field mapping method, including:

[0006] Obtain fields to be matched;

[0007] Match the fields to be matched by using a pre-trained semantic model and a semantic rule library to obtain a matching result;

[0008] Determine corresponding fields according to the matching result, and give a correction suggestion when there are conflict problems between the corresponding fields to obtain a corrected matching result.

[0009] Optionally, matching the fields to be matched by using a pre-trained semantic model and a semantic rule library to obtain a matching result includes:

[0010] Match the fields to be matched by using the pre-trained semantic model to obtain a first matching result of successful matching and unmatched fields of failed matching;

[0011] Match the unmatched fields by using the semantic rule library to obtain a second matching result.

[0012] Optionally, matching the fields to be matched by using the pre-trained semantic model to obtain a first matching result of successful matching and unmatched fields of failed matching includes:

[0013] Perform semantic analysis on the fields to be matched and extract semantic features;

[0014] Based on the pre-trained semantic model and the semantic features, calculate the semantic similarity between the fields;

[0015] Determine the first matching result and the unmatched fields according to the semantic similarity;

[0016] Optionally, use the semantic rule library to match the unmatched fields to obtain a second matching result, including:

[0017] Construct the semantic rule library based on the user's historical matching operations and the business-specific semantic library, and dynamically update the semantic rule library using machine learning;

[0018] For the unmatched fields, perform pattern matching using the semantic rule library to obtain the second matching result.

[0019] Optionally, it further includes:

[0020] Determine the corresponding fields according to the matching result, and when there are no conflict issues between the corresponding fields, update the semantic rule library based on the matching result.

[0021] Optionally, determine the corresponding fields according to the matching result, and when there are conflict issues between the corresponding fields, give a correction suggestion to obtain a corrected matching result, including:

[0022] Determine the corresponding fields based on the matching result;

[0023] Judge whether there are conflict issues between the corresponding fields, and the conflict issues at least include incompatible data types, inconsistent statistical calibers, and inconsistent time formats;

[0024] When the data types of the corresponding fields are incompatible, perform type conversion or complete the missing fields;

[0025] When the statistical calibers of the corresponding fields are inconsistent, provide a unified calculation formula for statistical caliber conversion;

[0026] When the time formats of the corresponding fields are inconsistent, convert them to a unified time format.

[0027] Optionally, obtain the fields to be matched, including:

[0028] Obtain the fields to be matched from each advertising multimedia channel.

[0029] The present invention also provides a field mapping device, including:

[0030] An acquisition module, configured to acquire fields to be matched;

[0031] A matching module, configured to match the fields to be matched by using a pre-trained semantic model and a semantic rule library, and obtain a matching result;

[0032] A correction module, configured to determine corresponding fields according to the matching result, and give a correction suggestion when there is a conflict problem between the corresponding fields, so as to obtain a corrected matching result.

[0033] The present invention further provides a field mapping device, including:

[0034] A memory, configured to store a computer program;

[0035] A processor, configured to implement the field mapping method as described above when executing the computer program.

[0036] The present invention further provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are loaded and executed by a processor, the field mapping method as described above is implemented.

[0037] It can be seen that the present invention acquires fields to be matched; matches the fields to be matched by using a pre-trained semantic model and a semantic rule library, and obtains a matching result; determines corresponding fields according to the matching result, and gives a correction suggestion when there is a conflict problem between the corresponding fields, so as to obtain a corrected matching result. The present invention performs comprehensive field matching based on a pre-trained semantic model and a semantic rule library, and also detects conflict problems of fields and performs intelligent correction, solves the problem of field heterogeneity in cross-channel data integration, realizes "zero-code" fast mapping, and improves the efficiency and accuracy of data standardization.

[0038] In addition, the present invention further provides a field mapping method, device, equipment and computer-readable storage medium, which also have the above beneficial effects. Description of the Drawings

[0039] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained according to the provided drawings without creative efforts.

[0040] Figure 1 It is a flowchart of a field mapping method provided by an embodiment of the present invention;

[0041] Figure 2A flowchart of a field mapping method provided by an embodiment of the present invention;

[0042] Figure 3 A structural schematic diagram of a field mapping device provided by an embodiment of the present invention;

[0043] Figure 4 A structural schematic diagram of a field mapping device provided by an embodiment of the present invention. Detailed implementation manners

[0044] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are only some of the embodiments of the present invention, rather than all of them. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0045] Please refer to Figure 1 , Figure 1 A flowchart of a field mapping method provided by an embodiment of the present invention. The method may include:

[0046] S101: Obtain fields to be matched.

[0047] The execution subject of this embodiment is a terminal. This embodiment does not limit the type of the terminal, as long as it can complete the operations of the field mapping method. It should be noted that the field matching and field mapping in this embodiment have the same meaning. The fields to be matched in this embodiment may be fields on a single advertising multimedia platform; or may also be fields on multiple advertising multimedia platforms.

[0048] S102: Match the fields to be matched by using a pre-trained semantic model and a semantic rule library to obtain a matching result.

[0049] This embodiment adopts two field matching methods, such as a pre-trained semantic model and a semantic rule base. This embodiment uses the pre-trained semantic model to automatically identify regular fields and complete the corresponding relationships of regular fields; based on the dynamic rule base, complex fields can be automatically identified and the corresponding relationships of complex fields can be completed. The present invention realizes all-round field matching through field matching at two levels. The semantic rule base in this embodiment is a dynamic semantic rule base constructed based on the user's historical matching operations, where "dynamic" means that the semantic rule base can be continuously updated to store the latest matching relationships in the semantic rule base, such as continuously and dynamically updating the semantic rule base through historical data + user habits + user adjustments. Among them, regular fields refer to those that meet the high-confidence matching conditions of the pre-trained semantic model (semantic similarity ≥ threshold θ1 and there is a prior of standardized mapping), such as common advertising fields like exposure count, click count, consumption, etc. Complex fields refer to those with cross-channel statistical caliber ambiguities and involve non-standard fields generated dynamically, such as user-defined fields, interaction values, conversion counts, etc.

[0050] Further, the above-mentioned method of using the pre-trained semantic model and the semantic rule base to match the fields to be matched to obtain a matching result specifically may include the following steps:

[0051] Step 21: Use the pre-trained semantic model to match the fields to be matched to obtain a first matching result of successful matching and unmatched fields of failed matching;

[0052] Step 22: Use the semantic rule base to match the unmatched fields to obtain a second matching result.

[0053] This embodiment first automatically identifies the corresponding relationships of more than 90% of the regular fields through the pre-trained semantic model; then constructs a dynamic rule base based on the user's historical operations to provide an intelligent recommended matching solution for the remaining complex fields, thereby realizing all-round field mapping.

[0054] Further, the above-mentioned method of using the pre-trained semantic model to match the fields to be matched to obtain a first matching result of successful matching and unmatched fields of failed matching specifically may include the following steps:

[0055] Step 211: Perform semantic analysis on the fields to be matched to extract semantic features;

[0056] Step 212: Calculate the semantic similarity between fields based on the pre-trained semantic model and the semantic features;

[0057] Step 213: Determine the first matching result and the unmatched fields of failed matching according to the semantic similarity.

[0058] In this embodiment, semantic analysis is performed on the field names in the advertising multi-channel data sources to extract semantic features; based on the pre-trained semantic model, the semantic similarity between the field names of different channels is calculated; according to the semantic similarity, the corresponding relationships of more than 90% of the conventional fields are automatically matched.

[0059] For example, for the field name "Click volume", its semantic feature "Number of clicks" is extracted; for the field name "Clicks", its semantic feature "Number of clicks" is extracted; for the field name "Conversion volume", its semantic feature "Number of successful conversions" is extracted; for the field name "Conversions", its semantic feature "Number of successful conversions" is extracted. The semantic similarity between "Click volume" and "Clicks" is calculated to be 0.95. The semantic similarity between "Conversion volume" and "Conversions" is calculated to be 0.98. The field names with a semantic similarity greater than 0.9 are regarded as corresponding relationships, so as to automatically match "Click volume" and "Clicks" as corresponding fields, and automatically match "Conversion volume" and "Conversions" as corresponding fields. Or, the field names with a semantic similarity greater than 0.95 are regarded as corresponding relationships, so as to automatically match "Conversion volume" and "Conversions" as corresponding fields.

[0060] Furthermore, the above-mentioned use of the semantic rule library to match the unmatched fields to obtain the second matching result may specifically include the following steps:

[0061] Step 221: Construct a semantic rule library based on the user's historical matching operations and the business-specific semantic library, and dynamically update the semantic rule library using machine learning;

[0062] Step 222: For the unmatched fields, perform pattern matching using the semantic rule library to obtain the second matching result.

[0063] This embodiment records the user's historical matching operations on complex fields, constructs a dynamic semantic rule library; for new complex fields, performs pattern matching based on the dynamic semantic rule library, and provides an intelligent recommended matching solution.

[0064] For example, if the user matches "Exposure" with "Impressions" as corresponding fields, the corresponding relationship is recorded in the dynamic semantic rule library. When encountering a new field "Display Count", based on the corresponding relationship between "Exposure" and "Impressions" in the dynamic semantic rule library, it is recommended to match "Display Count" with "Impressions" as corresponding fields. Another example, if the user matches "Cost" with "Cost" as corresponding fields, the corresponding relationship is recorded in the dynamic semantic rule library. When encountering a new field "Advertising Expenses", based on the corresponding relationship between "Cost" and "Cost" in the dynamic semantic rule library, it is recommended to match "Advertising Expenses" with "Cost" as corresponding fields.

[0065] S103: Determine the corresponding fields according to the matching result, and give a correction suggestion when there are conflict problems between the corresponding fields, and obtain the corrected matching result.

[0066] After the above two-layer matching in this embodiment, it is also necessary to determine whether there are conflict problems such as incompatible data types and inconsistent statistical calibers, and for the detected conflict problems, give intelligent correction suggestions, such as automatically completing missing fields and unifying time format conversion. The conflict problems are not limited in this embodiment. Exemplarily, conflict problems include incompatible data types, inconsistent statistical calibers, and inconsistent time types, etc.

[0067] For example, it is detected that the "Click Count" field is numeric, while the "Clicks" field is character type, then there is a conflict of incompatible data types. It is detected that the "Exposure" field uses the display count statistical caliber, while the "Impressions" field uses the video exposure statistical caliber, then there is a conflict of inconsistent statistical calibers. It is detected that the "Start Time" field is in date format, while the "StartTime" field is in timestamp format, then there is a conflict of inconsistent time formats. Another example, it is detected that the "Conversion Count" field is numeric, while the "Conversions" field is also numeric, then there is no conflict of incompatible data types. It is detected that the "Conversion Count" field uses the last click attribution model, while the "Conversions" field uses the linear attribution model, then there is a conflict of inconsistent statistical calibers. It is detected that the "Start Time" field is in date format, while the "StartTime" field is also in date format, then there is no conflict of inconsistent time formats.

[0068] Further, the above steps of determining the corresponding fields according to the matching result, and giving a correction suggestion when there are conflict problems between the corresponding fields, and obtaining the corrected matching result can specifically include the following steps:

[0069] Step 31: Determine the corresponding fields based on the matching result;

[0070] Step 32: Determine whether there are conflict issues between corresponding fields. The conflict issues include at least incompatible data types, inconsistent statistical calibers, and inconsistent time formats.

[0071] Step 33: When the data types of the corresponding fields are incompatible, perform type conversion or complete missing fields.

[0072] Step 34: When the statistical calibers of the corresponding fields are inconsistent, provide a unified calculation formula for statistical caliber conversion.

[0073] Step 35: When the time formats of the corresponding fields are inconsistent, convert them to a unified time format.

[0074] In this embodiment, it is detected whether the data types of fields from different channels are compatible, such as numeric type and character type; whether the statistical calibers of fields from different channels are consistent, such as differences in exposure / conversion attribution models; and whether the time formats of fields from different channels are consistent, such as timestamps and date formats. For fields with incompatible data types, type conversion is automatically performed or missing fields are completed; for fields with inconsistent statistical calibers, a unified calculation formula is provided for conversion; for fields with inconsistent time formats, they are automatically converted to a unified time format.

[0075] For example, convert the character-type "Clicks (number of clicks)" field to a numeric type, or complete the missing numeric-type "number of clicks" field; provide a formula to convert the video exposure statistical caliber to the display count statistical caliber; convert the "StartTime (start time)" field in timestamp format to date format; it is detected that the data types of the "conversion volume" and "Conversions (conversion)" fields are compatible, so there is no need to perform type conversion or complete missing fields; provide a formula to convert the linear attribution model to the last click attribution model to unify the statistical calibers of the "conversion volume" and "Conversions" fields; it is detected that the time formats of the "start time" and "StartTime" fields are consistent, so there is no need to perform time format conversion.

[0076] Furthermore, the above method may further include the following steps:

[0077] Determine the corresponding fields according to the matching result, and when there are no conflict issues between the corresponding fields, update the semantic rule library based on the matching result.

[0078] Specifically, even if there are no conflicts in the corresponding fields in the current matching result, the rule library can be updated using the recommended solutions in this matching process, and the system dynamically optimizes the recommendation strategy through reinforcement learning.

[0079] This embodiment can further provide a corresponding visualization interface based on the above method. This interface supports visual drag-and-drop interaction, that is, it supports batch field mapping and one-click conflict repair. For example, batch mapping of more than 10 fields, and intelligent recognition of semantic conflicts, highlighting them to guide users to confirm or modify. This visualization interface can be a two-column visualization layout, supporting drag-and-drop mapping of source fields and target fields. For example, when mapping fields on multiple advertising multimedia platforms to one platform, there is a one-to-many relationship. For example, the "click count" fields of advertising multimedia platform 1, advertising multimedia platform 2, and advertising multimedia platform 3 need to be mapped to the "click count" of advertising multimedia platform 4. Therefore, a two-column field configuration box design is required, and the one-to-many mapping of fields is completed by dragging and connecting lines.

[0080] Applying the field mapping method provided by the embodiment of the present invention, by obtaining the fields to be matched; using a pre-trained semantic model and a semantic rule library to match the fields to be matched to obtain a matching result; determining the corresponding fields according to the matching result, and giving a correction suggestion when there are conflict problems between the corresponding fields to obtain a corrected matching result. The present invention conducts comprehensive field matching based on a pre-trained semantic model and a semantic rule library, and also detects conflict problems in fields and performs intelligent correction, solves the problem of field heterogeneity in cross-channel data integration, realizes "zero-code" fast mapping, and improves the efficiency and accuracy of data standardization. Moreover, the single-channel field mapping time of this method is shortened from 30 minutes to ≤5 minutes, the manual error rate is reduced by 90%, improving the field mapping efficiency; more than 90% of the conventional field correspondence relationships are automatically recognized by the pre-trained semantic model, improving the field matching accuracy; a conflict detection algorithm is set up to give real-time warnings for problems such as incompatible data types and inconsistent statistical calibers, and give intelligent correction suggestions, with a conflict detection coverage rate of 100% and the statistical caliber consistency improved to more than 99%, improving the data quality; a dynamic rule library is constructed based on the user's historical matching operations, supporting dynamic addition of the rule library, and the new channel access cost is reduced by 70%, effectively solving the problems that the existing technology cannot support dynamic addition of the rule library and the new channel access cost is high, enhancing the scalability; reducing 75% of the technical dependence, non-technical users can independently complete 90% of the mapping tasks, effectively solving the problems that the existing technology requires a large number of technical personnel to participate and the cost is high, and greatly reducing the cost.

[0081] It should be further explained that, through human-computer interaction optimization, the data cleaning work that originally required code writing is converted into a graphical configuration operation, which increases the feasibility of non-technical personnel completing complex field mapping tasks from 18% to 79%. That is, the field mapping method of this embodiment can also provide an advertising multi-channel data field mapping user interface. Through visual comparison, configuration, and intelligent connection and prompts, the field matching time is shortened from an average of 3 minutes to 45 seconds, thereby improving the degree of automation, reducing the human error rate, and improving configuration efficiency. Moreover, this method can not only be used in the advertising attribution data background, but can also be widely used in other scenarios. For example:

[0082] (1) Advertising agencies: In cross-platform advertising scenarios, agencies often need to integrate multi-channel advertising data. Advertisers often need to filter by multiple dimensions such as platform type, advertising time period, and advertising format (e.g., viewing creative materials with ROI (Return on Investment) > 2% for information flow advertising during Q3). Traditional methods require repeated logging into different platforms to export data. This method can provide a targeted combination query function and establish a unified query portal. That is, by connecting the market APIs (market application programming interfaces) of various media channels, multimedia channel data can be obtained regularly, and the scattered advertising data fields can be intelligently mapped to achieve cross-platform data comparison and analysis. Customized reports covering core indicators such as click-through rate and conversion cost can be generated in 5 minutes, saving 80% of data processing time compared to traditional methods. Among them, targeted combination query refers to the use of a triple technical architecture of semantic mapping layer, logic conversion layer, and dynamic compilation layer to transform the multi-channel query logic that originally required manual processing into an automated configuration process, thereby increasing the work efficiency of advertising operators by 5-8 times (based on 300 real business scenario test data).

[0083] (2) Enterprise self-built advertising platform: When an enterprise needs to connect to emerging channels such as grass-planting notes on Channel 1 and lifestyle accounts on Channel 2, it often faces the problem of incompatible data formats (e.g., the interaction value on Channel 1 = the number of collections + the number of comments, while traditional platforms only count the number of clicks). The marketing team needs to quickly verify the advertising effectiveness of the new channels, but the traditional development mode requires designing a query module separately for each new channel. For example, an enterprise's self-built advertising will gradually connect to new channels, such as Channel 1, Channel 2, etc. If it is necessary to verify the advertising effectiveness of the new channels, it is necessary to combine the effectiveness of order types, jointly calculate the ROI, and then compare it with the ROI of other channels. The traditional mode is to develop a script to query data, such as ROI, for this channel separately, and code maintenance is required each time. Through the dynamic field adaptation technology of this method, non-standard indicators such as "interaction volume" and "exposure UV" can be automatically recognized, and data standardization can be achieved through visual rule configuration. For example, after configuring the conversion formula of "redemption rate = the number of coupon codes used / the number of coupons issued" for Channel 2, it can be immediately compared horizontally with the ROI indicators of other channels. Among them, visual rule configuration means that instead of configuring field mappings in the code, visual configuration is directly performed on the function page, and the field names, quantities, and mapping relationships can all be displayed on the page.

[0084] (3) Third-party data tools: Considering the weak technical capabilities of small and medium-sized customers, data service providers can use the intelligent mapping engine provided by this method to automatically associate business indicators such as "sales amount" with the original fields of different systems (such as order_amount (order amount) in ERP (Enterprise Resource Planning) and deal_total (total transaction amount) in CRM (Customer Relationship Management)). When a user sets the query condition of "high-net-worth customers in 2023" (customer unit price ≥ 500 and repeat purchase ≥ 3 times), the system automatically penetrates and associates multi-table data such as membership level and order flow. Compared with the traditional SQL (Structured Query Language) query method, the usage threshold is reduced by 90%. The actual measurement of a certain SAAS tool shows that the proportion of customers independently creating complex queries has increased from 17% to 63%.

[0085] To better understand this solution, reference can be made to Figure 2 , Figure 2 which is a flow example diagram of a field mapping method provided by an embodiment of the present invention, including:

[0086] Step 1: Automatically identify the corresponding relationship of conventional fields based on a pre-trained semantic model;

[0087] Step 2: Build a dynamic semantic rule library based on the user's historical matching operations, and provide an intelligent recommendation matching solution for complex fields;

[0088] Step 3: Determine whether there are conflict problems such as data type incompatibility, statistical caliber inconsistency, and time type inconsistency. If not, return to Step 2; if so, execute Step 4;

[0089] Step 4: For the detected conflict problems, give intelligent correction suggestions, such as automatically completing missing fields and unifying time format conversion.

[0090] Next, the field mapping device provided by the embodiment of the present invention will be introduced. The field mapping device described below can be correspondingly referred to the field mapping method described above.

[0091] Specifically, please refer to Figure 3 , Figure 3 which is a schematic structural diagram of a field mapping device provided by an embodiment of the present invention, and may include:

[0092] An acquisition module 100, configured to acquire fields to be matched;

[0093] A matching module 200, configured to match the fields to be matched by using a pre-trained semantic model and a semantic rule library to obtain a matching result;

[0094] A correction module 300, configured to determine corresponding fields according to the matching result, and give a correction suggestion when there are conflict problems between the corresponding fields to obtain a corrected matching result.

[0095] Based on the above embodiment, the matching module 200 may include:

[0096] A first matching unit, configured to match the fields to be matched by using the pre-trained semantic model to obtain a first matching result of successful matching and unmatched fields of failed matching;

[0097] A second matching unit, configured to match the unmatched fields by using the semantic rule library to obtain a second matching result.

[0098] Based on the above embodiment, the first matching unit may include:

[0099] A semantic analysis subunit, configured to perform semantic analysis on the fields to be matched and extract semantic features;

[0100] A calculation subunit, configured to calculate the semantic similarity between fields based on the pre-trained semantic model and the semantic features;

[0101] A first matching subunit, configured to determine the first matching result and the unmatched fields according to the semantic similarity.

[0102] Based on the above embodiments, the second matching unit may include:

[0103] A construction subunit, configured to construct the semantic rule library according to the user's historical matching operations and the business-specific semantic library, and dynamically update the semantic rule library by using machine learning;

[0104] A second matching subunit, configured to perform pattern matching on the unmatched fields by using the semantic rule library to obtain the second matching result.

[0105] Based on the above embodiments, the field mapping device may further include:

[0106] An update module, configured to determine the corresponding fields according to the matching result, and when there is no conflict problem between the corresponding fields, update the semantic rule library based on the matching result.

[0107] Based on the above embodiments, the correction module 300 may include:

[0108] A corresponding field unit, configured to determine the corresponding fields based on the matching result;

[0109] A conflict judgment unit, configured to judge whether there is a conflict problem between the corresponding fields, where the conflict problem at least includes incompatible data types, inconsistent statistical calibers, and inconsistent time formats;

[0110] A first correction unit, configured to perform type conversion or complement missing fields when the data types of the corresponding fields are incompatible;

[0111] A second correction unit, configured to provide a unified calculation formula for statistical caliber conversion when the statistical calibers of the corresponding fields are inconsistent;

[0112] A third correction unit, configured to convert to a unified time format when the time formats of the corresponding fields are inconsistent.

[0113] Based on the above embodiments, the obtaining module 100 may include:

[0114] An obtaining unit, configured to obtain the fields to be matched from each advertising multimedia channel.

[0115] It should be noted that, without affecting the logic, the order of the modules and units in the above field mapping device can be changed before and after.

[0116] Applying the field mapping method provided by the embodiments of the present invention, a obtaining module 100 is used to obtain fields to be matched; a matching module 200 is used to match the fields to be matched by using a pre-trained semantic model and a semantic rule library to obtain a matching result; a correction module 300 is used to determine corresponding fields according to the matching result, and give a correction suggestion when there is a conflict problem between the corresponding fields, so as to obtain a corrected matching result. The present invention performs comprehensive field matching based on a pre-trained semantic model and a semantic rule library, and also detects conflict problems of fields and performs intelligent correction, solves the problem of field heterogeneity in cross-channel data integration, realizes "zero-code" fast mapping, and improves the efficiency and accuracy of data standardization.

[0117] Next, a field mapping device provided by the embodiments of the present invention will be introduced. The field mapping device described below can be correspondingly referred to the field mapping method described above.

[0118] Please refer to Figure 4 , Figure 4 which is a schematic structural diagram of a field mapping device provided by the embodiments of the present invention, and may include:

[0119] A memory 10, which is used to store computer programs;

[0120] A processor 20, which is used to execute the computer program to implement the above-mentioned field mapping method.

[0121] The memory 10, the processor 20, and the communication interface 31 all complete communication with each other through a communication bus 32.

[0122] In the embodiments of the present invention, the memory 10 is used to store one or more programs. The program may include program codes, and the program codes include computer operation instructions. In the embodiments of the present invention, the memory 10 may store programs for implementing the following functions:

[0123] Obtain fields to be matched;

[0124] Match the fields to be matched by using a pre-trained semantic model and a semantic rule library to obtain a matching result;

[0125] Determine corresponding fields according to the matching result, and give a correction suggestion when there is a conflict problem between the corresponding fields, so as to obtain a corrected matching result.

[0126] In a possible implementation manner, the memory 10 may include a program storage area and a data storage area. Among them, the program storage area may store an operating system and application programs required for at least one function, etc.; the data storage area may store data created during the use process.

[0127] In addition, the memory 10 may include a read-only memory and a random access memory, and provide instructions and data to the processor. A part of the memory may also include NVRAM. The memory stores an operating system and operation instructions, executable modules, or data structures, or subsets thereof, or extended sets thereof, where the operation instructions may include various operation instructions for implementing various operations. The operating system may include various system programs for implementing various basic tasks and processing hardware-based tasks.

[0128] The processor 20 may be a central processing unit (CPU), an application-specific integrated circuit, a digital signal processor, a field programmable gate array, or other programmable logic devices. The processor 20 may be a microprocessor or any conventional processor, etc. The processor 20 may call the program stored in the memory 10.

[0129] The communication interface 31 may be an interface of a communication module for connecting to other devices or systems.

[0130] Of course, it should be noted that Figure 4 the structure shown does not constitute a limitation on the field mapping device in the embodiments of the present invention. In practical applications, the field mapping device may include more or fewer components than Figure 4 those shown, or combine certain components.

[0131] Next, the computer-readable storage medium provided by the embodiments of the present invention will be introduced. The computer-readable storage medium described below can be correspondingly referred to the field mapping method described above.

[0132] The present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned field mapping method are implemented.

[0133] The computer-readable storage medium may include various media that can store program codes, such as a USB flash drive, a portable hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc.

[0134] In this specification, the various embodiments are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The same or similar parts among the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple. For the relevant parts, reference can be made to the description in the method part.

[0135] Those skilled in the art may further realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described according to functions in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0136] Finally, it should also be noted that in this document, relationships such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or elements inherent to such process, method, article or device.

[0137] The above has introduced in detail a field mapping method, apparatus, device, and computer-readable storage medium provided by the present invention. Specific examples are used herein to elaborate on the principles and implementation manners of the present invention. The description of the above embodiments is only used to help understand the method and its core idea of the present invention; at the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims

1. A field mapping method, characterized in that, Including: Obtain the fields to be matched; Match the fields to be matched using a pre-trained semantic model and a semantic rule library to obtain a matching result; Determine the corresponding fields according to the matching result, and give a correction suggestion when there are conflict problems between the corresponding fields to obtain a corrected matching result.

2. The field mapping method according to claim 1, wherein Match the fields to be matched using a pre-trained semantic model and a semantic rule library to obtain a matching result, including: Match the fields to be matched using the pre-trained semantic model to obtain a first matching result of successful matching and unmatched fields of failed matching; Match the unmatched fields using the semantic rule library to obtain a second matching result.

3. The field mapping method according to claim 2, wherein Match the fields to be matched using the pre-trained semantic model to obtain a first matching result of successful matching and unmatched fields of failed matching, including: Perform semantic analysis on the fields to be matched to extract semantic features; Calculate the semantic similarity between fields based on the pre-trained semantic model and the semantic features; Determine the first matching result and the unmatched fields according to the semantic similarity.

4. The field mapping method according to claim 2, wherein Match the unmatched fields using the semantic rule library to obtain a second matching result, including: Construct the semantic rule library according to the user's historical matching operations and the business-specific semantic library, and dynamically update the semantic rule library using machine learning; For the unmatched fields, perform pattern matching using the semantic rule library to obtain the second matching result.

5. The field mapping method according to claim 1, wherein Also including: Determine the corresponding fields according to the matching result, and when there are no conflict problems between the corresponding fields, update the semantic rule library based on the matching result.

6. The field mapping method according to claim 1, wherein Determine the corresponding fields according to the matching result, and give a correction suggestion when there are conflict problems between the corresponding fields to obtain a corrected matching result, including: Determine the corresponding fields based on the matching result; Judge whether there are conflict problems between the corresponding fields, and the conflict problems at least include incompatible data types, inconsistent statistical calibers, and inconsistent time formats; When the data types of the corresponding fields are incompatible, perform type conversion or complete missing fields; When the statistical calibers of the corresponding fields are inconsistent, provide a unified calculation formula for statistical caliber conversion; When the time formats of the corresponding fields are inconsistent, convert them to a unified time format.

7. The field mapping method according to claim 1, characterized in that, Obtain the fields to be matched, including: Obtain the fields to be matched from each advertising multimedia channel.

8. A field mapping device, characterized in that, Including: An acquisition module for obtaining the fields to be matched; A matching module for matching the fields to be matched using a pre-trained semantic model and a semantic rule library to obtain a matching result; A correction module for determining the corresponding fields according to the matching result, and giving a correction suggestion when there are conflict problems between the corresponding fields to obtain a corrected matching result.

9. A field mapping device, characterized in that, Including: A memory for storing computer programs; A processor for implementing the field mapping method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are loaded and executed by a processor, the field mapping method described in any one of claims 1 to 7 is implemented.

Images