CN117807282A - Service data processing method, device, electronic equipment and readable storage medium - Google Patents

Abstract

This application provides a business data processing method, device, electronic equipment and readable storage medium, relating to the field of data processing. In this method, the retrieval asset data is first obtained; in response to the selection operation of the preset target marking model, the retrieval asset data is marked using the target marking model to obtain the marking result; according to the marking result, the preset target marking model is used to mark the retrieval asset data. Make the first recommendation based on the recommendation strategy and obtain relevant recommendation results, which include recommended data and tags corresponding to the recommended data; obtain a preset asset marking list, which includes multiple asset data to be marked and each to-be-marked asset data. Mark the marking tag corresponding to the asset data; obtain the matching result based on the recommended data, the tag corresponding to the recommended data, the asset data to be marked, and the marking tag; perform a second recommendation based on the matching result and the retrieved asset data, and obtain the recommended asset data , able to recommend accurate asset data to users.

Description

Business data processing methods, devices, electronic equipment and readable storage media

Technical field

This application relates to the field of data processing technology, specifically to a business data processing method, device, electronic equipment and readable storage medium.

Background technique

The asset data of an enterprise is data that the enterprise legally owns or controls. It is usually divided into different categories according to different properties. For example, it can be divided into different asset data such as products, markets, orders, costs, revenues, services, channels, etc. according to attributes. Asset data inventory and intelligent labeling can directly or indirectly bring economic benefits and provide new development opportunities for enterprises. Intelligent recommendation is an application scenario of asset data inventory and intelligent labeling of enterprises, which recommends appropriate asset data to users and realizes the value of enterprise assets. Intelligent recommendation recommends the recommended content that users want through user input query data. However, when users query data, they are not very clear about the search results and usually enter some related content. However, these related contents are relatively vague and cannot recommend accurate asset data to users.

Contents of the invention

This application provides a business data processing method, device, electronic device and readable storage medium, which can recommend accurate asset data to users.

The technical solution of the embodiment of the present application is as follows:

In a first aspect, embodiments of the present application provide a business data processing method, which method includes:

Get retrieval asset data;

In response to the selection operation of the preset target marking model, use the target marking model to mark the retrieved asset data to obtain a marking result;

According to the labeling result, a first recommendation is performed using a preset recommendation strategy to obtain a relevant recommendation result, wherein the relevant recommendation result includes the recommended data and a label corresponding to the recommended data;

Obtain a preset asset marking list, which includes a plurality of asset data to be marked and marking tags corresponding to each of the asset data to be marked;

Obtain a matching result based on the recommended data, the tag corresponding to the recommended data, the asset data to be marked and the marking tag;

A second recommendation is performed based on the matching result and the retrieved asset data to obtain recommended asset data.

In the above technical solution, the retrieval asset data is first obtained, and the retrieval asset data is the retrieval content input by the user. The retrieval asset data is obtained to provide data support for subsequent recommendation of accurate asset data to the user; in response to the preset target marking model Select the operation and use the target marking model to mark the retrieved asset data to obtain the marking result. First, mark the retrieved asset data. The obtained marking result indicates the category to which it belongs, which is conducive to accurate recommendation; according to the marking result, use The preset recommendation strategy is used for the first recommendation, and relevant recommendation results are obtained. The relevant recommendation results include recommended data and labels corresponding to the recommended data. First, the recommendation strategy is used to make rough recommendations, which is beneficial to improving the accuracy of the recommendation; the preset asset marking is obtained. List, the asset marking list includes multiple asset data to be marked and marking tags corresponding to each asset data to be marked, providing data support for subsequent accurate recommendations; based on the recommended data, the tags corresponding to the recommended data, and the asset data to be marked and marking tags to obtain matching results; perform a second recommendation based on the matching results and retrieved asset data to obtain recommended asset data. By performing the second recommendation, more accurate recommended asset data can be obtained.

In some embodiments of the present application, before obtaining the preset asset marking list, the method further includes:

Obtain multiple asset data to be marked, preprocess each of the asset data to be marked, and obtain multiple preprocessed asset data;

Use the target marking model to mark each of the preprocessed asset data to obtain a marking label;

The asset marking list is obtained according to each of the asset data to be marked and the marking tag corresponding to each of the asset data to be marked.

In the above technical solution, by preprocessing the asset data to be marked, that is, performing data screening and data cleaning, it can not only reduce the amount of calculation, but also improve the marking accuracy. Then use the target marking model to mark the pre-processed asset data to obtain marking labels. The asset data to be marked and the marking labels form an asset marking list, which is conducive to subsequent matching based on the asset marking list to recommend accurate asset data.

In some embodiments of the present application, before labeling each of the pre-processed asset data using the target labeling model to obtain a labeling label, the method further includes:

Obtain an asset sample data set, where the asset sample data set includes multiple asset sample data and marking labels corresponding to each of the asset sample data;

Preprocess each of the asset sample data to obtain multiple preprocessed sample data;

Performing a first sample balancing process on each of the preprocessed sample data to obtain a training data set;

Perform a second sample equalization process on each of the preprocessed sample data to obtain a verification data set;

Use the training data set to adjust parameters of the preset first initial marking model to obtain a second initial marking model;

The second initial marking model is verified using the verification data set, and when the accuracy is greater than the preset accuracy threshold, the target marking model is obtained.

In the above technical solution, the generalization of the model can be improved by preprocessing the asset sample data set, that is, performing data screening and data cleaning. Then perform sample balancing processing to obtain a training set and a verification set, use the training set to adjust parameters, and use the verification set for verification to obtain a target marking model, which has good generalization and robustness.

In some embodiments of the present application, the first sample equalization process is performed on each of the preprocessed sample data to obtain a training data set, including:

Perform deduplication processing on the preprocessed sample data according to categories, field Chinese names and field English names to obtain the first deduplicated sample data;

Select the top N categories from the first deduplication sample data to obtain selected sample data, where N is an integer greater than or equal to 2;

Deduplicate the selected sample data according to categories and field Chinese names to obtain the second deduplicated sample data;

Remove the categories whose number does not exceed the preset number in the second deduplication sample data to obtain the training data set;

Performing second sample equalization processing on each of the preprocessed sample data to obtain a verification data set, including:

Categories whose number does not exceed a preset number in the preprocessed sample data are removed to obtain the verification data set.

In the above technical solution, deduplication is first carried out according to categories, field Chinese names and field English names, and then deduplication is carried out according to categories and field Chinese names to achieve effective deduplication and avoid duplicate records. Reduce the impact of invalid data on model accuracy by removing categories that do not exceed a preset number in the data.

In some embodiments of the present application, before labeling the retrieved asset data using the target labeling model to obtain a labeling result, the method further includes:

Perform vector processing on the retrieved asset data to obtain an asset data vector;

Perform data dimension upgrading processing on the retrieved asset data to obtain a multi-dimensional data vector;

The asset data vector and the multi-dimensional data vector are fused and then input into the target marking model.

In the above technical solution, by performing data dimensionality upgrading, the associated data of the retrieved asset data can be increased to clarify the recommendation target, and then data fusion is performed to increase the data representation ability, which is conducive to improving the accuracy of the model output.

In some embodiments of the present application, obtaining a matching result based on the recommended data, the tag corresponding to the recommended data, the asset data to be marked, and the marking tag includes:

Classify the tags corresponding to the recommended data and the marked tags into categories, calculate the similarity between the recommended data and the asset data to be marked, and filter out data whose tag categories are consistent and whose similarity is less than a preset threshold. , to obtain the matching result.

In the above technical solution, accurate recommendations are made based on categories and similarities, and matching results are obtained, which is beneficial to subsequent recommendations of more accurate asset data for users based on the matching results.

In some embodiments of the present application, the recommendation strategy is at least one of quantity limitation, matching degree limitation, and combination limitation.

In the above technical solution, quantity limitation, matching degree limitation, and combination limitation have different advantages. Users can choose corresponding strategies for recommendation according to their needs, so that the recommended content meets the user's needs.

In a second aspect, an embodiment of the present application provides a service data processing device, the device comprising:

The first data acquisition module is used to obtain and retrieve asset data;

A marking module, configured to respond to a selection operation of a preset target marking model, use the target marking model to mark the retrieved asset data, and obtain a marking result;

A first recommendation module, configured to make a first recommendation based on the labeling result and using a preset recommendation strategy to obtain a relevant recommendation result, wherein the relevant recommendation result includes the recommended data and a label corresponding to the recommended data;

The second data acquisition module is used to obtain a preset asset marking list. The asset marking list includes a plurality of asset data to be marked and marking tags corresponding to each of the asset data to be marked;

A matching module, configured to obtain a matching result based on the recommended data, the tag corresponding to the recommended data, the asset data to be marked, and the marking tag;

The second recommendation module is used to make a second recommendation based on the matching result and the retrieved asset data to obtain recommended asset data.

In a third aspect, embodiments of the present application provide an electronic device, including a processor, a memory, a user interface, and a network interface. The memory is used to store instructions, and the user interface and network interface are used to communicate with other devices. The processor is configured to execute instructions stored in the memory, so that the electronic device executes any one of the methods provided in the first aspect.

In a fourth aspect, embodiments of the present application provide a computer-readable storage medium that stores instructions. When the instructions are executed, any one of the methods provided in the first aspect is executed. method.

To sum up, one or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

1. Since the retrieved asset data is obtained and the target marking model is used to mark the retrieved asset data, the first recommendation is made based on the marking results and relevant recommendation results are obtained, thereby filtering out asset data similar to the retrieved asset data. Then, based on the relevant recommendation results and the asset marking list, the matching results are obtained, and the second recommendation is made based on the matching results and the retrieved asset data, and the technical means of recommending asset data are obtained. Therefore, it effectively solves the problem that the related content input in the related technology is relatively vague and unclear. This leads to the inability to recommend accurate asset data to users. The retrieved asset data is marked through the target marking model, the first recommendation is made based on the marking results, the asset data similar to the retrieved asset data is screened out, and the second recommendation is made based on the asset marking list. Multi-layer recommendation can provide Users recommend accurate asset data.

2. Using data preprocessing and sample equalization to process data can improve the generalization and robustness of the model.

3. Perform vector processing and data dimensionality processing on the data, and then input it into the model, which can improve the accuracy of the training model.

Description of drawings

Figure 1 is one of the flow diagrams of a business data processing method provided by an embodiment of the present application;

Figure 2 is a second schematic flowchart of a business data processing method provided by an embodiment of the present application;

Figure 3 is the third schematic flowchart of a business data processing method provided by an embodiment of the present application;

Figure 4 is the fourth schematic flowchart of a business data processing method provided by an embodiment of the present application;

Figure 5 is a schematic diagram of label division of a business data processing method provided by an embodiment of the present application;

Figure 6 is a schematic flowchart of a sub-step of step S930 in Figure 4;

Figure 7 is a schematic diagram of sample equalization of a business data processing method provided by an embodiment of the present application;

FIG8 is a schematic diagram of the overall flow of a business data processing method provided by an embodiment of the present application;

Figure 9 is a schematic structural diagram of a business data processing device provided by an embodiment of the present application;

Figure 10 is a schematic structural diagram of an electronic device provided by an embodiment of the present application.

Detailed ways

In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this specification. Obviously, the described The embodiments are only some of the embodiments of this application, not all of them.

In the description of the embodiments of this application, words such as "for example" or "for example" are used to represent examples, illustrations or illustrations. Any embodiment or design described as "such as" or "for example" in the embodiments of the present application shall not be construed as being preferred or advantageous over other embodiments or designs. Rather, the use of words such as "for example" or "for example" is intended to present the concept in a concrete manner.

In the description of the embodiments of this application, the term “plurality” means two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Therefore, features defined as "first" and "second" may explicitly or implicitly include one or more of these features. The terms “including,” “includes,” “having,” and variations thereof all mean “including but not limited to,” unless otherwise specifically emphasized.

Embodiments of the present application provide a business data processing method, device, electronic device and readable storage medium. The business data processing method first obtains retrieval asset data. The retrieval asset data is the retrieval content input by the user, and the retrieval asset data is obtained for subsequent processing. Users recommend accurate asset data to provide data support; in response to the selection operation of the preset target marking model, use the target marking model to mark the retrieved asset data, and obtain the marking results. First, mark the retrieved asset data. , the obtained labeling result indicates the category to which it belongs, which is conducive to accurate recommendation; according to the labeling result, the preset recommendation strategy is used to make the first recommendation, and relevant recommendation results are obtained. The relevant recommendation results include recommended data and labels corresponding to the recommended data. First Using the recommendation strategy to make rough recommendations is helpful to improve the accuracy of recommendations; obtain a preset asset marking list, which includes multiple asset data to be marked and the marking tags corresponding to each asset data to be marked, which provides the basis for subsequent Provide data support for accurate recommendations; obtain matching results based on recommended data, tags corresponding to recommended data, asset data to be marked, and marking tags; make a second recommendation based on matching results and retrieved asset data, obtain recommended asset data, and perform the third Second recommendation, more accurate recommended asset data can be obtained. Compared with the prior art in which the input related content is relatively vague and cannot recommend accurate asset data to users, the embodiment of the present application marks the retrieved asset data through a target marking model, and makes the first recommendation based on the marking results. , filter out asset data similar to the retrieved asset data, and then make a second recommendation based on the asset marking list. Through multi-layer recommendation, accurate asset data can be recommended to users.

It should be noted that this business data processing method is applied in the banking field, commercial asset recommendation and other fields, and can also be applied to other asset recommendation fields. First, through asset inventory and marking management, the compiled asset marking list is used for intelligent recommendations, which can recommend accurate asset data to users.

The technical solutions provided by the embodiments of this application will be further described below with reference to the accompanying drawings, taking the banking industry as an example.

Referring to Figure 1, Figure 1 is a flow chart of a business data processing method provided in an embodiment of the present application. The business data processing method is applied to a business data processing device, and the business data processing method is executed by a processor in an electronic device or a readable storage medium. The business data processing method includes steps S100, S200, S300, S400, S500, and S600.

Step S100: Obtain retrieval asset data.

In one embodiment, retrieving asset data is a brief summary or specific description of the data that the user wants to obtain. Retrieving asset data may include loan policy search, banking financial management search, deposit policy search, etc. The user can search through the preset Enter the retrieval asset data on the web page or APP, and obtain the retrieval asset data through the preset interface, which is conducive to subsequent recommendations based on the retrieval asset data and provides data support. Among them, the default interface includes the input() function.

Step S200, in response to the selection operation of the preset target labeling model, the retrieved asset data is labeled using the target labeling model to obtain a labeling result.

In one embodiment, the user selects a target labeling model on a web page. Different models are recommended in different ways. The selected target labeling model can make the recommended data more accurate. In response to the selection operation of the preset target labeling model, the retrieved asset data is labeled using the target labeling model to obtain a labeling result. The labeling result is a specific subcategory obtained after the data is divided according to attributes or categories. The category division can be based on industry classification standards. Since the target labeling model is pre-trained, the target labeling model can accurately mark the category of the retrieved asset data, which is conducive to subsequent recommendations based on the labeling results. The target labeling model can be a BERT model, a variant of the BERT model, or other deep learning models, as long as it can meet the accurate labeling requirements, which will not be elaborated here.

As shown in FIG2 , before the target labeling model is used to label the retrieved asset data and the labeling result is obtained, the business data processing method further includes but is not limited to the following steps:

Step S710: Perform vector processing on the retrieved asset data to obtain an asset data vector.

In one embodiment, before the retrieved asset data is input into the target marking model, vector processing is first performed on the retrieved asset data. Specifically, the retrieved asset data is vectorized using a preset data representation algorithm to obtain an asset data vector. Through data vectorization, the retrieved asset data is converted into a data type that can be processed by the target marking model, which is beneficial to improving computing efficiency.

Step S720, perform data dimensionality upgrading processing on the retrieved asset data to obtain a multi-dimensional data vector.

In one embodiment, data dimension upgrading is to expand relevant data for the data. As shown in Figure 8, data dimension upgrading for user data can be expanded into roles, departments, focused businesses, historical retrieval records, jurisdictional systems, etc., and can be based on The pre-established knowledge graph performs data dimensionality processing on the retrieved asset data to obtain multi-dimensional data vectors. The search scope of retrieving asset data can be increased by performing data dimensionality processing, which is beneficial to improving the accuracy of recommendations. The knowledge graph can reflect the correlation between each data, so that the knowledge graph can be used to quickly increase the data dimension. The knowledge graph is pre-constructed according to specific business scenarios. Its construction process is similar to other knowledge graph construction, so I will not go into details here.

Step S730: The asset data vector and the multi-dimensional data vector are fused and input into the target labeling model.

In one embodiment, the asset data vector and the multi-dimensional data vector are fused. The asset data vector and the multi-dimensional data vector can be used for feature connection, and then input into the target marking model. The target marking model performs marking and can output More accurate marking categories, thereby improving the accuracy of recommendations.

Step S300, based on the labeling result, a first recommendation is performed using a preset recommendation strategy to obtain a relevant recommendation result, which includes the recommended data and a label corresponding to the recommended data.

In one embodiment, the first recommendation is a rough recommendation based on the retrieved asset data. Data of different categories from the marked tags are first removed, and only content related to the retrieved asset data is recommended. According to the marking results obtained in step S200, the preset recommendation strategy is used to make the first recommendation, and relevant recommendation results are obtained, which is conducive to subsequent advancement based on the relevant recommendation results, thereby recommending accurate asset data to the user. The relevant recommendation results include recommended data and tags corresponding to the recommended data. The tags corresponding to the recommended data and the marking results belong to the same category. The recommended data has a strong correlation with the retrieved asset data, which is represented by the matching degree. The higher the matching degree, the higher the correlation. High means stronger correlation.

In another embodiment, the recommendation strategy is at least one of quantity limitation, matching degree limitation, and combination limitation. When the recommendation strategy is limited in quantity, the recommendation matching degree ranks the top A data, and the recommendation strategy recommends faster. Among them, a is an integer greater than 0 and less than 100, a can be 5, etc., and can be adjusted according to the situation, which will not be described here. When the recommendation strategy is limited by matching degree, data whose matching degree is greater than the preset correlation threshold is recommended. This recommendation strategy has faster recommendation speed and higher recommendation accuracy. Among them, the correlation threshold can be 70%, etc., and can be adjusted according to the situation, which will not be described here. When the recommendation strategy is a combination restriction, the recommendation matching degree is greater than the preset association threshold, and the matching degree ranks the top A data, the recommendation speed of this recommendation strategy is slow, and the recommendation accuracy is the highest. For example, the default matching degree arrangement is 10 pieces of data, and the correlation threshold is 70%. The different strategies are: the quantity is limited to the top 5 items in the matching degree ranking, the matching degree is limited to a matching degree greater than 70%, and the combination limit matches The top 5 items with a degree of ≥70% and the highest matching degree.

Step S400: Obtain a preset asset marking list. The asset marking list includes multiple asset data to be marked and marking tags corresponding to each asset data to be marked.

In one embodiment, the preset asset marking list is obtained by marking multiple asset data to be marked, and each asset data to be marked and the marking tag corresponding to each asset data to be marked are A tag forms a record in the asset tagging list. Obtain the preset asset marking list through the preset data acquisition interface to provide data preparation for subsequent recommendations of more accurate asset data. The preset data acquisition interface can be the read() function.

As shown in Figure 3, before obtaining the preset asset marking list, the business data processing method also includes but is not limited to the following steps:

Step S810, obtaining a plurality of asset data to be marked, preprocessing each asset data to be marked, and obtaining a plurality of preprocessed asset data.

In one embodiment, the asset data to be marked includes different data assets such as products, markets, orders, costs, revenue, services, channels, etc. Multiple asset data to be marked can be obtained through crawler technology, or through the analysis of banking services. Historical data is compiled and obtained. Then organize and save multiple asset data to be marked into scv, txt or other table formats, use the data reading interface to obtain multiple asset data to be marked, and provide data support for subsequent marking to obtain the asset marking list.

In one embodiment, each asset data to be marked is preprocessed, and the preprocessing includes data cleaning and data vectorization. Data cleaning performs data screening and removal of messy data on multiple asset data to be marked. Data screening is to select data columns that affect the marking classification of the asset data to be marked, such as the Chinese name of the field, the English name of the field, the Chinese name of the table, the English name of the table, the field type and the field length, etc., and filter out important field data, which can reduce the amount of calculation while ensuring accuracy. Removing messy data can be removing blank lines, garbled characters, notes and duplicate data. Among them, removing duplicate data can remove duplicate record rows according to fields such as category, field Chinese name and field English name, and only retain one row of records in the same data. By performing data preprocessing to obtain multiple preprocessed asset data, the amount of calculation can be reduced and the marking accuracy can be improved.

Step S820: Mark each preprocessed asset data using the target marking model to obtain marking labels.

In one embodiment, the target marking model is a trained model. The target marking model is used to mark the preprocessed asset data obtained in step S810. The preprocessed asset data is directly input into the target marking model. The target marking model Outputting the marking tag will facilitate subsequent construction of the asset marking list.

Step S830: Obtain an asset marking list based on each asset data to be marked and the marking tag corresponding to each asset data to be marked.

In one embodiment, each asset data to be marked and the marking tag corresponding to the asset data to be marked are regarded as one record, and multiple asset data to be marked and their corresponding marking tags constitute multiple records, and multiple records are formed. Records form an asset marking list. Since the multiple asset data to be marked are data with different attributes in the industry, the asset marking list includes relatively rich asset data and its marking tags, which is conducive to subsequent use of the asset marking list to match relevant recommendation results and recommend accurate information to users. Asset data.

As shown in FIG4 , before using the target labeling model to label each pre-processed asset data to obtain a labeling label, the business data processing method further includes but is not limited to the following steps:

Step S910: Obtain an asset sample data set. The asset sample data set includes multiple asset sample data and marking labels corresponding to each asset sample data.

In one embodiment, the asset sample data set includes multiple asset sample data. The multiple asset sample data includes different data assets such as products, markets, orders, costs, revenue, services, channels, etc. Multiple asset samples can be obtained through crawler technology. data, and then organize and save multiple asset sample data into scv, txt or other table formats. The asset sample data set also includes marking labels corresponding to each asset sample data. The marking labels are obtained by manually labeling multiple asset sample data and are stored in a file corresponding to each asset sample data. Use the data reading interface to obtain asset sample data sets to provide data support for subsequent model training.

It should be noted that the asset sample data is manually labeled, and the labeling rules are divided according to the standards published by the banking industry. The asset sample data is matched with the corresponding sub-categories to obtain the labeling labels. As shown in Figure 5, the categories include: general for participants, general for individuals, general for organizations, customer information-basic information, risk information-credit risk, management personality-individual customers, corporate customers, financial peer customers, employee information, institutional information, merchant information-basic information, terminal information, etc. Each category corresponds to a corresponding level, for example, merchant information-basic information corresponds to level 2. When the asset sample data is input as merchant information, the sub-category under the merchant information is further determined, and the basic information sub-category is then determined. The label of the asset sample data is expressed as merchant information-basic information-/-2.

Step S920: Preprocess each asset sample data to obtain multiple preprocessed sample data.

In one embodiment, each asset sample data is preprocessed, and the preprocessing includes data cleaning and data vectorization. Data cleaning performs data screening on multiple asset sample data, removes cluttered data, etc. Data screening is to select data columns that have an impact on labeling and classification of asset sample data, such as field Chinese name, field English name, table Chinese name, and table English name. , field type and field length, etc., filtering out important field data can reduce the amount of calculation while ensuring accuracy. Removing cluttered data can include removing blank lines, garbled characters, notes, duplicate data, etc. Among them, removing duplicate data can remove duplicate record rows according to categories, field Chinese names, field English names and other fields, and only retain one row of records in the same data. By performing data preprocessing to obtain multiple preprocessed sample data, the amount of calculation can be reduced and the accuracy of model marking can be improved.

Step S930: Perform a first sample equalization process on each preprocessed sample data to obtain a training data set.

As shown in FIG6 , performing a first sample balancing process on each preprocessed sample data to obtain a training data set includes but is not limited to the following steps:

Step S931: Perform deduplication processing on the preprocessed sample data according to the category, field Chinese name and field English name to obtain the first deduplication sample data.

In some possible embodiments of this application, based on the preprocessed sample data obtained in step S920, deduplication processing is performed according to categories, field Chinese names and field English names. For example, if the categories, Chinese field names, and English names of multiple pieces of preprocessed sample data are the same, the multiple pieces of preprocessed sample data are duplicate data. The duplicate data will be deleted and only one preprocessed sample of the same data will be guaranteed. data to obtain the first deduplication sample data, which is beneficial to subsequent training sets based on the first deduplication sample data.

Step S932: Select N categories ranked first in number from the first deduplication sample data to obtain selected sample data, where N is an integer greater than or equal to 2.

In some possible embodiments of the present application, N is an integer greater than or equal to 2, and N is not greater than 10. Select the N categories with the highest number from the first deduplication sample data. When N is 2, the two categories with the highest number are selected to obtain the selected sample data, which is beneficial to subsequent secondary deduplication based on the selected sample data. Heavy. It should be noted that this category is a major category, and the major category also includes multiple sub-categories.

Step S933: Deduplicate the selected sample data according to categories and field Chinese names to obtain second deduplicated sample data.

In some possible embodiments of the present application, since the Chinese name of the field can be translated into different English, based on the selected sample obtained in step S932, deduplication processing is performed again according to the category and the Chinese name of the field to obtain the second deduplication sample data. This can avoid the existence of data with the same meaning in English and thus form duplicate data, and the second deduplication sample data has no duplicate data.

Step S934: Remove categories whose number does not exceed a preset number in the second deduplication sample data to obtain a training data set.

In some possible embodiments of the present application, the second deduplication sample data is obtained according to step S933, and the number of the second deduplication sample data is counted. In each small category, the number of the second deduplication sample data that is removed does not exceed the predetermined number. Set the number of categories. The preset number can be 10 or 20. Remove the categories whose number is less than or equal to 10 in the second deduplication sample data to obtain a training data set. Constructing a training data set through the above steps will help improve the accuracy of the model.

Step S940: Perform a second sample equalization process on each preprocessed sample data to obtain a verification data set.

In one embodiment, a second sample equalization process is performed on each preprocessed sample data to obtain a verification data set, which specifically includes but is not limited to: removing categories whose number does not exceed a preset number in the preprocessed sample data. The preset number can be It can be 10 items or 20 items. Remove the categories whose number is less than or equal to 10 items in the preprocessed sample data to obtain the verification data set, which will help improve the accuracy of the model.

As shown in Figure 7, assuming that there are 5789 pieces of preprocessed sample data, the preprocessed sample data is deduplicated according to the category, field Chinese name and field English name, and then the top 2 numbers are selected from the deduplicated data. categories; deduplicate the selected data according to categories and field Chinese names, and obtain the deduplicated data volume of 2252 items; then remove categories with no more than 10 items in the data to obtain a training data set with 2250 data items. Categories with no more than 10 items were removed from the preprocessed sample data to obtain a verification data set with 5780 data items.

Step S950: Use the training data set to adjust parameters of the preset first initial marking model to obtain a second initial marking model.

In one embodiment, the training data set includes preprocessed sample data without duplication and whose data volume meets the preset quantity, and the labeling label corresponding to the preprocessed sample data. The preprocessed sample data and the labeling label are used to adjust the parameters of the preset first initial labeling model to obtain the second initial labeling model, which is conducive to the subsequent acquisition of the target labeling model. The first initial labeling model can be a BERT model, a variant of the BERT model, or other deep learning models, as long as it can meet the accurate labeling requirements, which will not be elaborated here; the model parameter adjustment is the parameter adjustment method of the BERT model, which will not be elaborated here.

Step S960: Use the verification data set to verify the second initial marking model. If the accuracy is greater than the preset accuracy threshold, the target marking model is obtained.

In one embodiment, the training data set includes pre-processed sample data that meets a preset number and does not include the marking labels corresponding to the pre-processed sample data. The verification data set is used to verify the second initial marking model. In terms of accuracy, If it is greater than the preset accuracy threshold, it indicates that the generalization of the model is good, the target marking model is obtained, and the retrieval asset data is marked with high accuracy.

In another embodiment, when the accuracy is less than or equal to the preset accuracy threshold, the training data set is subjected to data dimensionality processing, and the first initial marking model is adjusted using the dimensionally increased training data set. parameters so that the accuracy is greater than the preset accuracy threshold.

In another embodiment, when the bank provides new sample data, the sample data has no corresponding labeling tags, and the number of categories in the new sample data does not exceed the preset number is removed, and the target labeling model is tested, and the result is as high as 82%. The accuracy of this target marking model has good generalization.

Step S500: Obtain matching results based on recommended data, tags corresponding to the recommended data, asset data to be marked, and marking tags.

In one embodiment, the matching results are obtained based on the recommended data, the tags corresponding to the recommended data, the asset data to be marked, and the marking tags. Specifically, the matching results include but are not limited to: first, classifying the tags corresponding to the recommended data and the marking tags. , use the preset classification algorithm to determine whether the tags corresponding to the recommended data and the marked tags are in the same subcategory, and use the similarity algorithm to calculate the similarity between the recommended data and the asset data to be marked, and filter out the tag categories that are consistent and similar. For data that is less than the preset threshold, the matching results are obtained, which is conducive to subsequent accurate asset data recommendations based on the matching results. The classification algorithm can be a support vector machine or a deep learning algorithm, etc.; the similarity algorithm can be Euclidean distance, character similarity algorithm, etc., which will not be described here.

Step S600: Perform a second recommendation based on the matching result and the retrieved asset data to obtain recommended asset data.

In one embodiment, the second recommendation is an accurate recommendation based on the matching results. Among the data of the same category, asset data that better meets the user's needs is recommended. It is advantageous for the matching result obtained in step S500 to have high similarity and the same label, and highly match the retrieved asset data. The recommended data ranked first is recommended to obtain the recommended asset data. After multiple screenings, the recommended asset data is more in line with the relevant requirements for retrieving asset data, so that accurate asset data can be recommended to users.

It should be noted that the corresponding tags in the recommended asset data can be used as tags for retrieving asset data, and the retrieved asset data and its corresponding tags are added to the asset marking list to increase the amount of data in the asset marking list, which can improve Recommended accuracy.

As shown in Figure 8, Figure 8 shows an overall flow diagram of a business data processing method provided by an embodiment of the present application. First, the retrieval asset data is obtained, and in response to the selection operation of the preset target marking model, the Retrieve asset data for vector processing, and perform dimensionality upgrading processing on the retrieved asset data, and fuse the processed data into the target marking model for marking processing to obtain the marking results. The obtained marking results indicate the category to which they belong, which is conducive to accuracy Recommendation; based on the marking results, use one of the preset quantity restrictions, matching degree restrictions, and combination restrictions to make the first recommendation, and obtain relevant recommendation results. The relevant recommendation results include recommended data and labels corresponding to the recommended data. First use The recommendation strategy makes rough recommendations, which is helpful to improve the accuracy of recommendations.

In one embodiment, multiple asset data to be marked are obtained, each asset data to be marked is preprocessed to obtain multiple preprocessed asset data, and each preprocessed asset data is marked using a target marking model to obtain Labeling; confirm and publish the marked data, inventory the asset data, and achieve effective management of assets. According to the confirmed and published asset data to be marked and the corresponding marking of each asset data to be marked, Tag to get the asset tag list. Obtain the preset asset marking list. The asset marking list includes multiple asset data to be marked and the marking tags corresponding to each asset data to be marked; based on the recommended data, the tags corresponding to the recommended data, the asset data to be marked and Mark the tag to get the matching result; make a second recommendation based on the matching result and the retrieved asset data to get the recommended asset data. By making the second recommendation, you can get more accurate recommended asset data.

As shown in FIG9 , an embodiment of the present application provides a business data processing device 100. The business data processing device 100 acquires retrieved asset data through a first data acquisition module 110. The retrieved asset data is the retrieval content input by the user. The retrieved asset data is acquired to provide data support for subsequently recommending accurate asset data to the user. The tagging module 120 is used to respond to the selection operation of the preset target tagging model, and the retrieved asset data is tagged using the target tagging model to obtain the tagging result. The retrieved asset data is first tagged, and the obtained tagging result indicates the category to which it belongs, which is conducive to accurate recommendation. Then, the first recommendation module 130 is used to make a first recommendation based on the tagging result using a preset recommendation strategy. Relevant recommendation results are obtained, which include recommended data and labels corresponding to the recommended data. A rough recommendation is first performed using a recommendation strategy, which is beneficial to improving the accuracy of the recommendation. A second data acquisition module 140 is used to obtain a preset asset labeling list, which includes multiple asset data to be labeled and labeling labels corresponding to each asset data to be labeled, to provide data support for subsequent accurate recommendations. The matching module 150 is then used to obtain matching results based on the recommended data, the labels corresponding to the recommended data, the asset data to be labeled, and the labeling labels. Finally, the second recommendation module 160 is used to perform a second recommendation based on the matching results and the retrieved asset data to obtain recommended asset data, so that relatively accurate recommended asset data can be obtained.

It should be noted that the first data acquisition module 110 is connected to the marking module 120, the marking module 120 is connected to the first recommendation module 130, the first recommendation module 130 is connected to the second data acquisition module 140, and the second data acquisition module 140 The matching module 150 is connected to the second recommendation module 160 . The above business data processing method is applied to the business data processing device 100. The business data processing device 100 marks the retrieved asset data through the target marking model, makes first recommendations based on the marking results, and filters out asset data similar to the retrieved asset data. , and then make a second recommendation based on the asset marking list. Through multi-layer recommendation, accurate asset data can be recommended to users.

It should also be noted that when the device provided in the above embodiment implements its functions, only the division of the above functional modules is used as an example. In actual application, the above functions can be allocated to different functional modules according to needs, that is, The internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be described again here.

The present application also discloses an electronic device. Referring to FIG10 , FIG10 is a schematic diagram of the structure of an electronic device disclosed in an embodiment of the present application. The electronic device 500 may include: at least one processor 501 , at least one network interface 504 , a user interface 503 , a memory 505 , and at least one communication bus 502 .

Among them, the communication bus 502 is used to realize connection communication between these components.

Among them, the user interface 503 may include a display screen (Display) and a camera (Camera), and the optional user interface 503 may also include a standard wired interface and a wireless interface.

Among them, the network interface 504 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface).

Among them, the processor 501 may include one or more processing cores. The processor 501 uses various interfaces and lines to connect various parts of the entire server, and executes the server by running or executing instructions, programs, code sets or instruction sets stored in the memory 505, and calling data stored in the memory 505. Various functions and processing data. Optionally, the processor 501 can use at least one of digital signal processing (Digital Signal Processing, DSP), field-programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). implemented in hardware form. The processor 501 may integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processor (Graphics Processing Unit, GPU), a modem, etc. Among them, the CPU mainly handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content that needs to be displayed on the display; and the modem is used to handle wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 501 and may be implemented by a separate chip.

Among them, the memory 505 may include a random access memory (RAM) or a read-only memory (Read-Only Memory). Optionally, the memory 505 includes a non-transitory computer-readable storage medium. The memory 505 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 505 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 505 may also be optionally at least one storage device located away from the aforementioned processor 501. Referring to Figure 10, the memory 505 as a computer storage medium may include an operating system, a network communication module, a user interface module, and an application program of a business data processing method.

In the electronic device 500 shown in FIG10 , the user interface 503 is mainly used to provide an input interface for the user and obtain the data input by the user; and the processor 501 can be used to call an application program storing a business data processing method in the memory 505, and when executed by one or more processors 501, the electronic device 500 executes one or more methods in the above-mentioned embodiments. It should be noted that for the aforementioned method embodiments, for the sake of simple description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the present application.

In the above embodiments, each embodiment is described with its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

Among the several implementations provided in this application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into Another system, or some features can be ignored, or not implemented. Another point is that the coupling or direct coupling or communication connection between each other shown or discussed may be through some service interfaces, and the indirect coupling or communication connection of the devices or units may be in electrical or other forms.

A unit described as a separate component may or may not be physically separate. A component shown as a unit may or may not be a physical unit, that is, it may be located in one place, or it may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present application can be integrated into one processing unit, each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated units can be implemented in the form of hardware or software functional units.

If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a memory and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the various embodiments of the present application. The aforementioned memory includes: various media that can store program codes, such as USB flash drives, mobile hard drives, magnetic disks or optical disks.

The above are only exemplary embodiments of the present disclosure and do not limit the scope of the present disclosure. That is to say, all equivalent changes and modifications made based on the teachings of this disclosure are still within the scope of this disclosure. Other embodiments of the present disclosure will readily occur to those skilled in the art, upon consideration of the specification and disclosure of practical truths.

This application is intended to cover any variations, uses, or adaptations of the disclosure that follow the general principles of the disclosure and include common knowledge or customary technical means in the technical field that are not described in the disclosure. . It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being defined by the following claims.

Claims (10)

1. A business data processing method, characterized in that the method includes: Get retrieval asset data; In response to the selection operation of the preset target marking model, use the target marking model to mark the retrieved asset data to obtain a marking result; According to the labeling result, a first recommendation is performed using a preset recommendation strategy to obtain a relevant recommendation result, wherein the relevant recommendation result includes the recommended data and a label corresponding to the recommended data; Obtaining a preset asset marking list, wherein the asset marking list includes a plurality of asset data to be marked and a marking label corresponding to each of the asset data to be marked; Obtain a matching result based on the recommended data, the tag corresponding to the recommended data, the asset data to be marked and the marking tag; A second recommendation is made based on the matching result and the retrieved asset data to obtain recommended asset data. 2. The method according to claim 1, characterized in that before obtaining the preset asset marking list, the method further includes: Obtain multiple asset data to be marked, preprocess each of the asset data to be marked, and obtain multiple preprocessed asset data; Using the target labeling model to label each of the pre-processed asset data to obtain a labeling label; The asset marking list is obtained according to each of the asset data to be marked and the marking tags corresponding to each of the asset data to be marked. 3. The method according to claim 2, characterized in that, before using the target marking model to mark each of the pre-processed asset data to obtain a marking label, the method further includes: Obtain an asset sample data set, where the asset sample data set includes multiple asset sample data and marking labels corresponding to each of the asset sample data; Preprocessing each of the asset sample data to obtain a plurality of preprocessed sample data; Performing a first sample balancing process on each of the preprocessed sample data to obtain a training data set; Perform a second sample equalization process on each of the preprocessed sample data to obtain a verification data set; Using the training data set to adjust the parameters of the preset first initial labeling model to obtain a second initial labeling model; The second initial marking model is verified using the verification data set, and when the accuracy is greater than the preset accuracy threshold, the target marking model is obtained. 4. The method according to claim 3, characterized in that, performing a first sample equalization process on each of the preprocessed sample data to obtain a training data set, including: Perform deduplication processing on the preprocessed sample data according to categories, field Chinese names and field English names to obtain the first deduplicated sample data; Selecting N categories with the largest number from the first deduplicated sample data to obtain selected sample data, where N is an integer greater than or equal to 2; Deduplicate the selected sample data according to categories and field Chinese names to obtain the second deduplicated sample data; Removing the categories whose number does not exceed a preset number in the second deduplication sample data to obtain the training data set; Performing second sample equalization processing on each of the preprocessed sample data to obtain a verification data set, including: Categories whose number does not exceed a preset number in the preprocessed sample data are removed to obtain the verification data set. 5. The method according to claim 1, characterized in that, before using the target marking model to mark the retrieved asset data to obtain a marking result, the method further includes: Perform vector processing on the retrieved asset data to obtain an asset data vector; Performing data dimensionality upgrading processing on the retrieved asset data to obtain a multi-dimensional data vector; The asset data vector and the multi-dimensional data vector are fused and then input into the target marking model. 6. The method according to claim 1, characterized in that the matching result is obtained based on the recommended data, the tag corresponding to the recommended data, the asset data to be marked and the marking tag, including : Classify the tags corresponding to the recommended data and the marked tags into categories, calculate the similarity between the recommended data and the asset data to be marked, and filter out data whose tag categories are consistent and whose similarity is less than a preset threshold. , to obtain the matching result. 7. The method according to claim 1 is characterized in that the recommendation strategy is at least one of quantity restriction, matching degree restriction, and combination restriction. 8. A business data processing device, characterized in that the device includes: The first data acquisition module (110) is used to acquire retrieval asset data; A marking module (120), configured to respond to a selection operation of a preset target marking model, use the target marking model to mark the retrieved asset data, and obtain a marking result; The first recommendation module (130) is configured to use a preset recommendation strategy to make a first recommendation based on the marking results, and obtain relevant recommendation results, where the relevant recommendation results include recommendation data and labels corresponding to the recommendation data; A second data acquisition module (140) is used to acquire a preset asset marking list, wherein the asset marking list includes a plurality of asset data to be marked and a marking label corresponding to each of the asset data to be marked; A matching module (150), configured to obtain a matching result based on the recommended data, the tag corresponding to the recommended data, the asset data to be marked, and the marking tag; The second recommendation module (160) is used to make a second recommendation based on the matching result and the retrieved asset data to obtain recommended asset data. 9. An electronic device, characterized in that it includes a processor (501), a memory (505), a user interface (503), a communication bus (502) and a network interface (504). The processor (501), the The memory (505), the user interface (503), and the network interface (504) are respectively connected to the communication bus (502). The memory (505) is used to store instructions, and the user interface (503) ) and network interface (504) are used to communicate with other devices, and the processor (501) is used to execute instructions stored in the memory (505), so that the electronic device (500) executes claims 1-7 any of the methods described. 10. A computer-readable storage medium, characterized in that the computer-readable storage medium stores instructions, and when the instructions are executed, the method according to any one of claims 1-7 is performed.