Index query method and device, equipment, storage medium and program product

Abstract

The invention provides an index query method and device, equipment, a storage medium and a program product, and relates to the field of financial science and technology or other related fields. The method comprises the steps that a to-be-queried business index and the search priority of the business index are obtained, and the search priority is related to the attribute of the business index; and searching in a target database according to the to-be-queried business index and the search priority to obtain candidate data. And according to a pre-trained similarity model, determining a matching degree between each candidate data and the to-be-queried business index, the similarity model being generated based on historical search record training of the target database. And if the first candidate data of which the matching degree is greater than or equal to a preset matching degree threshold exists, generating a query result of the to-be-queried business index according to the first candidate data, and outputting the query result. According to the method, the index query efficiency and accuracy are improved.

Description

Technical Field

[0001] The present application relates to the field of financial technology or other related fields, and in particular to an indicator query method, device, equipment, storage medium and program product. Background Art

[0002] With the advent of the big data era, enterprises have an increasing demand for data indicators, which are essential for business decision-making, operational monitoring, and market analysis. However, in actual operations, there are many challenges in finding and verifying the availability of the required data indicators. The traditional method relies on manual access to data indicator websites and querying them one by one, which is very inefficient. In addition, the diversity of data sources, differences in the timeliness of data tables, missing data table attribute information, and possible invalidation of data tables further increase the difficulty of obtaining data indicators. Even if the data table corresponding to the indicator is found, it is still necessary to perform secondary verification on the data query platform. The whole process is cumbersome and time-consuming.

[0003] Therefore, how to improve the efficiency of indicator query is an urgent problem to be solved. Summary of the invention

[0004] The present application provides an indicator query method, device, equipment, storage medium and program product to improve the efficiency of indicator query.

[0005] In a first aspect, the present application provides an indicator query method, comprising:

[0006] Acquire a business indicator to be queried, and a search priority of the business indicator, where the search priority is related to an attribute of the business indicator;

[0007] Searching in a target database to obtain candidate data according to the business indicator to be queried and the search priority;

[0008] Determining the matching degree between each candidate data and the business indicator to be queried according to a pre-trained similarity model, wherein the similarity model is generated and trained based on the historical search records of the target database;

[0009] If there is first candidate data whose matching degree is greater than or equal to a preset matching degree threshold, a query result of the business indicator to be queried is generated according to the first candidate data.

[0010] The query result is output.

[0011] In a second aspect, the present application provides an indicator query device, comprising:

[0012] An acquisition module, used to acquire the business indicator to be queried, and a search priority of the business indicator, wherein the search priority is related to an attribute of the business indicator;

[0013] A search module, used to search in a target database to obtain candidate data according to the business indicator to be queried and the search priority;

[0014] A processing module, used to determine the matching degree between each candidate data and the business indicator to be queried according to a pre-trained similarity model, wherein the similarity model is generated based on the training of the historical search records of the target database;

[0015] The control module is used to generate a query result of the business indicator to be queried according to the first candidate data if there is first candidate data with a matching degree greater than or equal to a preset matching degree threshold.

[0016] An output module is used to output the query result.

[0017] In a third aspect, the present application provides an electronic device, comprising: a processor, and a memory communicatively connected to the processor;

[0018] The memory stores computer-executable instructions;

[0019] The processor executes the computer-executable instructions stored in the memory to implement the method as described in the first aspect.

[0020] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement the method described in the first aspect.

[0021] In a fifth aspect, the present application provides a computer program product, which, when executed by a processor, is used to implement the method as described in the first aspect.

[0022] The indicator query method, device, equipment, storage medium and program product provided by the present application obtain the business indicator to be queried, and the search priority of the business indicator related to the attribute of the business indicator. According to the business indicator to be queried and the search priority, a search is performed in the target database to obtain candidate data. According to a pre-trained similarity model, the matching degree between each candidate data and the business indicator to be queried is determined. If there is a first candidate data whose matching degree is greater than or equal to a preset matching degree threshold, a query result of the business indicator to be queried is generated according to the first candidate data, and the query result is output, thereby improving the efficiency and accuracy of indicator query. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0024] Figure 1 A schematic diagram of a scenario for an indicator query provided in an embodiment of the present application;

[0025] Figure 2 A flowchart of an indicator query method provided in an embodiment of the present application;

[0026] Figure 3 A flowchart of another indicator query method provided in an embodiment of the present application;

[0027] Figure 4 A flowchart of another indicator query method provided in an embodiment of the present application;

[0028] Figure 5 A flowchart of another indicator query method provided in an embodiment of the present application;

[0029] Figure 6 A flowchart of another indicator query method provided in an embodiment of the present application;

[0030] Figure 7 A flowchart of another indicator query method provided in an embodiment of the present application;

[0031] Figure 8 A schematic diagram of the structure of an indicator query device provided in an embodiment of the present application;

[0032] Fig. 9 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0033] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0034] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0035] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with the relevant laws, regulations and standards of the relevant countries and regions, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0036] In addition, this application involves conducting big data analysis of user information (including but not limited to personal biometrics, identity data, consumption data, asset data, electronic terminal operation data, etc.), and using artificial intelligence technology to make automated decisions, and making technical solutions that have a significant impact on personal rights and interests based on the results of automated decisions. The application provides users with corresponding operation entrances for them to choose to agree or reject the results of automated decisions; if the user chooses to reject, the expert decision-making process will be entered.

[0037] It should be noted that the indicator query method, electronic device, storage medium and program product provided in this application can be used in the field of financial technology or other related fields, and can also be used in any field other than the field of financial technology or other related fields. The application field of the indicator query method, electronic device, storage medium and program product in this application is not limited.

[0038] With the advent of the big data era, enterprises' demand for data indicators has shown a trend of continuous growth. Data indicators play an indispensable and important role in many key links of enterprises, such as business decision-making, operation monitoring and market analysis. However, in terms of practical operations, the process of exploring and verifying the availability of required data indicators faces many challenges. The traditional operation method mainly relies on manual access to data indicator websites one by one, which is extremely inefficient. From the perspective of the characteristics of data sources, their diversity is extremely significant. Data from different sources have wide differences in structure, format and data quality. At the same time, the timeliness of data tables is also uneven, which means that the data in some data tables may not reflect the current actual situation in a timely manner. What's more difficult is that the attribute information of data tables is often missing, which makes enterprises lack sufficient basis when evaluating whether the data meets the needs. In addition, data tables may also have the problem of invalidation, that is, due to various reasons (such as updates and maintenance on the data source end or errors in the data transmission process, etc.), the data table cannot provide valid data normally.

[0039] Therefore, how to improve the efficiency of indicator query is an urgent problem to be solved.

[0040] In view of this, the present application provides an indicator query method, which obtains the business indicator to be queried and the search priority of the business indicator related to the attribute of the business indicator. According to the business indicator to be queried and the search priority, a search is performed in the target database to obtain candidate data. According to a pre-trained similarity model, the matching degree of each candidate data with the business indicator to be queried is determined. If there is a first candidate data whose matching degree is greater than or equal to a preset matching degree threshold, a query result of the business indicator to be queried is generated according to the first candidate data, and the query result is output, thereby improving the efficiency and accuracy of the indicator query.

[0041] Figure 1 A schematic diagram of a scenario of an indicator query provided in an embodiment of the present application. Figure 1 As shown, the scenario may include: a target database and an indicator query device.

[0042] The target database is a database used to query the index to be queried. The index query device is a user's terminal device, such as a mobile phone, a computer, a tablet computer, a bank teller, and other terminal devices.

[0043] The indicator query device uses Robotic Process Automation (RPA) technology to automatically perform an automatic search in the target database based on the business indicators to be queried and the search priority of the business indicators entered by the user, so as to obtain business indicator search results corresponding to the search priority of the business indicators and the business indicators to be queried.

[0044] The execution entity of the indicator query provided in this application can be Figure 1 The indicator query device mentioned herein, or the processing chip of the indicator query device, may also be software or program code for implementing the indicator query method.

[0045] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0046] Figure 2 A flowchart of an indicator query method provided in an embodiment of the present application is shown in FIG. Figure 2 As shown, the method may include:

[0047] S201. Obtain a business indicator to be queried and a search priority of the business indicator, where the search priority is related to an attribute of the business indicator.

[0048] The business indicator to be queried refers to the information content related to the specific business that the bank needs to query, such as the customer's credit score, account balance, loan application status, etc. The search priority is the query order weight set according to the attributes of the business indicator. For example, it may include at least one of priority level, asset type, timeliness, database type, branch opening rules, etc. For example, for business indicators related to high-risk transactions (such as large transfers), the search priority may be higher because the relevant information needs to be obtained quickly and accurately; while for general account history transaction query business indicators, the search priority may be lower.

[0049] Exemplarily, the search priorities may include: user tag && daily update && database 1 && full row sharing (priority 1), subject aggregation || data table && daily update && database 1 && full row sharing (priority 2), user tag && monthly update && database 2 && full row sharing (priority 3)

[0050] One possible implementation method is to receive query requests sent from different business departments within the bank (such as the credit department, the wealth management department, etc.), which contain clear business indicators to be queried. Assign a search priority to each business indicator according to predefined rules. These predefined rules are formulated based on the bank's long-term consideration of factors such as the importance, urgency, and risk level of various businesses. For example, for the business indicator of the credit department to inquire whether a customer has overdue loans, its search priority is set to high because it involves fund risk recovery; while for the business indicator of the wealth management department to inquire whether a customer has participated in a certain type of low-risk wealth management product, the search priority is set to low. The received business indicators to be queried and the corresponding search priorities are stored and managed for subsequent query operations. Optionally, the staff can manually add the identification (such as indicator name, indicator code) or description content of the business indicators to be queried, or batch import the list of business indicators to be queried.

[0051] Another possible implementation method is to determine the business type (such as risk control, customer service, etc.) to which the business indicator belongs based on the key information in the business indicator attributes. Different business types correspond to different default search priorities. For example, risk control business indicators have a higher search priority by default. When obtaining the business indicator to be queried, its business type information is also obtained to determine its search priority.

[0052] Optionally, for batch-imported business indicators to be queried, the files uploaded by the user can be parsed to accurately extract the business indicators to be queried in the files. In addition, the parsing results of the files can be previewed on the user interface, and the user can be prompted to modify the business indicators to be queried that do not conform to the naming rules specified by the system.

[0053] S202: Search the target database to obtain candidate data according to the business indicator to be queried and the search priority.

[0054] The target database in the bank system may be a database storing various types of bank business data, including a customer information database, a transaction record database, a financial product database, etc. At the same time, the target database may also be a website that can query the business indicators to be queried, and the websites corresponding to different business indicators to be queried may be the same or different. For example, for querying bank financial product information, the target database may be the bank's internal financial product database, or it may be a website section on the bank's official website that provides financial information query.

[0055] Specifically, the operation can be performed in order of search priority from high to low. If the search priority corresponds to database search: for the business indicator corresponding to the highest priority, the RPA technology is used to start the search process for the target database. For example, the RPA technology can be used to automatically open the database management tool according to the predefined process rules, locate the corresponding search portal according to the attributes of the business indicator, and then enter the business indicator to be queried into the search portal for search to obtain search results. If the search priority corresponds to website search: the website to be visited can be determined according to the indicator to be queried (for example, different business indicators correspond to different websites), and then the RPA technology can be used to simulate manual operations, set search conditions in the target database or website according to the search priority, and search with the business indicator to be queried as the search content to obtain search results.

[0056] S203: Determine the matching degree between each candidate data and the business indicator to be queried according to the pre-trained similarity model.

[0057] Among them, the similarity model is a model generated by training based on the historical search records of the target database, which is used to measure the similarity between candidate data and the business indicators to be queried. In the banking system, the similarity model can learn the association patterns between different business indicators and data based on a large number of historical search records.

[0058] Each candidate data and the business indicator to be queried may be input into a pre-trained similarity model in a prescribed format to determine the matching degree between each candidate data and the business indicator to be queried.

[0059] In one possible implementation, the similarity model calculates the matching score between each candidate data and the business indicator to be queried based on a predefined algorithm to reflect the similarity between each candidate data and the business indicator to be queried in terms of data content. The matching degree between each candidate data and the business indicator to be queried is determined based on the matching score.

[0060] In another possible implementation, the similarity model can calculate the matching score between each candidate data and the business indicator to be queried, and the matching score between the attribute of the candidate data and the search priority (to characterize the matching degree between the candidate data and the business indicator to be queried in terms of the attributes related to the search priority). Based on the matching score between each candidate data and the business indicator to be queried, and the matching score between the attribute of the candidate data and the search priority, the matching degree between each candidate data and the business indicator to be queried is comprehensively determined.

[0061] S204: If there is first candidate data whose matching degree is greater than or equal to a preset matching degree threshold, generate a query result of the business indicator to be queried according to the first candidate data.

[0062] The preset matching degree threshold is pre-set according to actual needs, and is used to determine whether the matching degree between the candidate data and the business indicator to be queried meets the expected requirements. In the banking system, the threshold can be set according to business needs and data accuracy requirements. For example, if the accuracy of the query result is required to be very high, the preset matching degree threshold can be set to a higher value (such as 0.8).

[0063] A possible implementation method is to traverse the matching degree of all candidate data with the business indicator to be queried, and compare all the matching degrees obtained with the preset matching degree threshold. If there are candidate data with a matching degree greater than or equal to the preset matching degree threshold, these candidate data are marked as first candidate data. Optionally, the first candidate data can be further sorted and screened according to business needs. For example, if there are multiple first candidate data that meet the conditions and there are some data conflicts, the optimal first candidate data is selected according to factors such as the update time of the data and the credibility of the data source, or multiple first candidate data are fused and processed to finally generate the query results of the business indicator to be queried.

[0064] Another possible implementation method is to use a sorting algorithm to sort all candidate data from high to low according to the matching degree, and to use the candidate data with a matching degree greater than or equal to a preset matching degree threshold as the first candidate data, and generate a query result based on it.

[0065] S205: Output the query results.

[0066] In one possible implementation, the generated query results may be packaged in a predetermined format. For example, if the query result is data about the customer account balance, it may be packaged in a standard format including information such as the customer account number, account balance, query time, etc. The packaged query result is then sent to the bank department that requested the query (such as a teller terminal, a customer manager terminal, etc.) for display, thereby completing the output operation.

[0067] Another possible implementation method is to directly convert the query results into a visual data format (such as tables, charts, reports, etc.), and then push the visualized query results to the bank's internal query system interface for display, thereby realizing the output of the query results.

[0068] The method provided by the embodiment of the present application obtains the business indicator to be queried, and the search priority of the business indicator related to the attribute of the business indicator. According to the business indicator to be queried and the search priority, a search is performed in the target database to obtain candidate data. According to a pre-trained similarity model, the matching degree of each candidate data and the business indicator to be queried is determined. If there is a first candidate data whose matching degree is greater than or equal to a preset matching degree threshold, a query result of the business indicator to be queried is generated according to the first candidate data, and the query result is output, thereby improving the efficiency and accuracy of the indicator query.

[0069] Next, how to determine the matching degree between each candidate data and the business indicator to be queried according to the pre-trained similarity model in the aforementioned step S203 is described in detail. Figure 3 A flowchart of another indicator query method provided in an embodiment of the present application. Figure 3 As shown, the aforementioned step S203 may specifically include:

[0070] S301. Input each candidate data and the business indicator to be queried into a similarity model to obtain a first matching score between each candidate data and the business indicator to be queried, and a second matching score between an attribute of each candidate data and a search priority.

[0071] The first match score is a quantitative score of the degree of match between the candidate data and the business indicator to be queried in terms of content, etc. This score is calculated by the similarity model. The higher the score, the more similar the two are in terms of data content. For example, when querying a customer's account balance business indicator, if the account balance data in the candidate data is very close to the account balance to be queried, then the first match score will be relatively high.

[0072] The second matching score represents a quantitative score of the degree of matching between the attributes of each candidate data and the search priority. Optionally, the second matching score can be calculated by a similarity model, or the matching score of the search priority can be set according to a preset priority order. When the second matching score sets the matching score of the search priority according to a preset priority order, for example, it can be assumed that there are n priorities. If the attributes of each candidate data match the first priority, the second matching score can be set to 1; if the attributes of each candidate data match the second priority, the second matching score can be set to n-1 / n; and so on, if the attributes of each candidate data match the nth priority, the second matching score can be set to 1 / n. It should be understood that the specific second matching score rule can be set according to actual needs. The above is only an example provided in the embodiment of the present application and is not a limitation.

[0073] Specifically, each candidate data and the business indicator to be queried can be sorted according to a pre-set input format. For example, the candidate data and the business indicator to be queried are converted into a specific data structure (such as a vector form), where the elements in the data structure can be key attribute values ​​in the business indicator (such as data type, update frequency, etc.) or metadata of the data (such as data source identifier, timestamp, etc.). Then the sorted candidate data and the business indicator to be queried are input into a pre-trained similarity model. Based on its predefined algorithm logic, the similarity model performs content matching analysis on the candidate data and the business indicator to be queried, and calculates the first matching score of each candidate data and the business indicator to be queried (for example, the similarity determined by calculating the semantic similarity between the candidate data and the business indicator to be queried can be used as the first matching score).

[0074] Among them, taking the similarity determined by calculating the semantic similarity between the candidate data and the business indicator to be queried as the first matching score as an example, the training process of the similarity model generated by training based on the historical search records of the target database can be implemented by the following sub-steps:

[0075] S3011. Obtain historical search records of the target database.

[0076] Among them, the historical search records include historical search business indicators and historical click results. The historical search records are the historical behavior records corresponding to the target database or website, the historical search business indicators are the search content entered by the staff when conducting historical searches on the target database or website, and the historical click results are the user click results in the search results corresponding to the search content.

[0077] Optionally, the historical search records can be obtained from the log database of the banking system, which is specially used to store historical search operations. By parsing the obtained log data, the records containing historical search business indicators and historical click results are screened out. For example, by identifying specific identifiers or fields, it is determined which data are historical search business indicators and which are the corresponding historical click results. The screened historical search business indicators and historical click results are sorted and stored to form a historical search record data set. Alternatively, the pre-stored historical search records can be directly obtained from the storage area of ​​the banking system that specifically stores historical search records according to a pre-set query strategy.

[0078] A training set and a test set of a similarity model are constructed based on historical search records to train the initial model and generate a trained similarity model. The similarity model may be, for example, a Deep Structured Semantic Models (DSSM) model or a BERT model.

[0079] S3012: Extract a first semantic feature vector of a historical search service indicator and a second semantic feature vector of a historical click result according to the historical search record.

[0080] The first semantic feature vector is a mathematical representation obtained after semantic analysis of the historical search business indicator, which can reflect the characteristics of the historical search business indicator in the semantic space. For example, for the historical search business indicator "high-yield financial products", its first semantic feature vector may contain numerical representations related to semantics such as "high yield" and "financial products". The second semantic feature vector is a mathematical representation obtained after semantic analysis of the historical click results. Similar to the first semantic feature vector, it can reflect the characteristics of the historical click results in the semantic space. For example, the historical click results are detailed information on a specific high-yield financial product, and its second semantic feature vector contains the semantic numerical representations of the various attributes of the financial product.

[0081] Each word or phrase in the historical search business indicator can be converted into a corresponding word vector through a pre-trained natural language processing (NLP) model, such as a word vector model (such as Word2Vec). Then, these word vectors are combined into a vector of fixed length according to certain combination rules (such as average pooling, weighted summation, etc.) as the first semantic feature vector of the historical search business indicator. For historical click results, the pre-trained NLP model is also used to convert the text content therein into a word vector, and then the second semantic feature vector is formed according to the same or similar combination rules.

[0082] S3013: Determine the similarity between the historical search service indicator and the historical click result based on the first semantic feature vector and the second semantic feature vector.

[0083] For example, the similarity between the first semantic feature vector and the second semantic feature vector can be determined by calculating the cosine similarity, Euclidean distance similarity, etc. between the first semantic feature vector and the second semantic feature vector, and the similarity can be used as the similarity between the historical search business indicator and the historical click result.

[0084] S3014: Train an initial model to generate a similarity model based on the first semantic feature vector, the second semantic feature vector, and the similarity between the historical search service indicators and the historical click results.

[0085] The first semantic feature vector, the second semantic feature vector, and the similarity of the historical search business indicators and the historical click results are used as training data. For the initial model of the neural network structure, the first semantic feature vector and the second semantic feature vector are used as the input of the input layer, and the similarity is used as the expected output of the output layer. During the model training process, the back propagation algorithm can be used to adjust the parameters of the model. According to the difference between the predicted result of the output layer and the expected output (i.e., similarity), the loss function (such as the mean square error loss function) is calculated, and then the parameters such as weights and biases in the model are adjusted according to the gradient back propagation of the loss function. After multiple rounds of iterative training, until the model converges, a similarity model is generated.

[0086] S302: Determine the matching degree between each candidate data and the business indicator to be queried according to the first matching score and the second matching score.

[0087] In a possible implementation, the sum of the first matching score and the second matching score may be directly used as the matching degree between each candidate data and the business indicator to be queried.

[0088] Another possible implementation method is to set corresponding weights for the first matching score and the second matching score, and obtain the matching degree between each candidate data and the business indicator to be queried by weighted calculation of the first matching score and the second matching score. For example, the weight of the first matching score can be set to 0.8, and the weight of the second matching score can be set to 0.2, then the score corresponding to the matching degree between each candidate data and the business indicator to be queried is equal to the first matching score*0.8+the second matching score*0.2.

[0089] The implementation method provided by the embodiment of the present application is to input each candidate data and the business indicator to be queried into a similarity model to obtain a first matching score between each candidate data and the business indicator to be queried, and a second matching score between the attribute of each candidate data and the search priority, and determine the matching degree of each candidate data and the business indicator to be queried according to the first matching score and the second matching score. This method comprehensively determines the matching degree of each candidate data and the business indicator to be queried by respectively determining the semantic similarity between each candidate data and the business indicator to be queried, and the priority similarity between the attribute of each candidate data and the search priority, and further improves the accuracy of the indicator query compared to the similarity between the candidate data and the business indicator to be queried only.

[0090] In the following, taking the case where the search priority includes at least two priorities as an example, how to search in the target database to obtain candidate data according to the business indicator to be queried and the search priority in the aforementioned step S202 is introduced. Figure 4 A flowchart of another indicator query method provided in an embodiment of the present application is shown below. Figure 4 As shown, the aforementioned step S202 may specifically include:

[0091] S401. Search in a target database according to the attributes of the business indicator corresponding to the first priority and the business indicator to be queried to obtain search results.

[0092] If the search result is empty, it indicates that it is necessary to switch to the next priority level for searching, and step S402 is executed; if the search result is not empty, it indicates that the corresponding candidate data has been searched, and step S403 is executed.

[0093] According to the attributes of the business indicators corresponding to the first priority, the attributes of the business indicators corresponding to the first priority are searched in the target database to determine whether there is candidate data that meets the attributes of the business indicators corresponding to the first priority. According to the results of this search, it is determined whether there is candidate data that meets the attributes of the business indicators corresponding to the first priority. If not, the search result is empty. If so, the search result is not empty, and candidate data that meets the attributes of the business indicators corresponding to the first priority is obtained.

[0094] S402: Search in the target database according to the attributes of the business indicator corresponding to the second priority and the business indicator to be queried to obtain search results.

[0095] The second priority level is located one priority level after the first priority level.

[0096] According to the order of search priority, each priority is searched in this order. If the business indicator to be queried is searched in the target database according to the attribute of the business indicator corresponding to the second priority level which is the next priority level after the first priority level, and the search result is not empty, the candidate data that meets the attribute of the business indicator corresponding to the second priority level is used as the search result; if the search result is empty, it means that the candidate data that meets the attribute of the business indicator corresponding to the second priority level cannot be searched, and the subsequent search priorities are searched in order according to the order of priority until the search result is not empty and then the search is stopped.

[0097] Optionally, if the corresponding candidate data cannot be found after searching all search priorities, relevant information such as "search failed" can be output to prompt the staff that the search has failed.

[0098] S403: Use the search results as candidate data.

[0099] The method provided in the embodiment of the present application searches the target database to obtain search results based on the attributes of the business indicators corresponding to the first priority and the business indicators to be queried. If the search result is empty, each priority is searched in this order until the search result is not empty. The search result at this time is used as candidate data, thereby further improving the completeness of the indicator query.

[0100] In one possible implementation, if there is no first candidate data with a matching degree greater than or equal to a preset matching degree threshold, the present application can also perform a supplementary search through the following steps to further improve the coverage of the indicator query. Figure 5 A flowchart of another indicator query method provided in an embodiment of the present application is shown below. Figure 5 As shown, the method may also include:

[0101] S501. Generate a database query instruction for the business indicator to be queried according to the business indicator to be queried and a preset template.

[0102] The database query instruction may be an SQL query instruction, a MySQL query instruction, etc. The specific type of the database query instruction is related to the type of database, and this application does not limit this. The preset template is a predefined query instruction framework that includes a fixed structure part of the query statement. Taking the SQL database as an example, the preset template may be, for example, a keyword (SELECT, FROM, WHERE, etc.) in the SQL query statement and some common conditional formats.

[0103] Optionally, preset template mapping relationships corresponding to different types of business indicators may be pre-stored. When the business indicator to be queried is received, the corresponding preset template is found in the mapping relationship according to the type of the business indicator (such as customer information query, transaction record query, etc.). For example, the specific content of the business indicator to be queried may be analyzed, and the key information therein may be extracted and filled into the corresponding position of the preset template, thereby generating a database query instruction.

[0104] Optionally, you can use a template engine tool that can dynamically generate query instructions based on the passed in parameters. First, load the preset template into the template engine. By parsing the business indicator to be queried, the key information obtained from the parsing (such as query conditions, target fields, etc.) is passed as parameters to the template engine. The template engine converts the preset template into a complete database query instruction based on these parameters. For example, if the business indicator to be queried is to query transaction records within a specific date, the template engine generates the corresponding query instruction based on the preset transaction record query template and the passed in date parameter.

[0105] S502: According to the business indicator to be queried, determine a data source for querying the business indicator to be queried through a database query instruction.

[0106] Optionally, the data source for querying the business indicator to be queried through the database query instruction can be determined by using a data source mapping table that records the correspondence between different types of business indicators and data sources, as well as the business indicator to be queried. For example, the data source mapping table may include the customer basic information business indicator corresponding to the customer information database, the transaction record business indicator corresponding to the transaction database, etc. The corresponding data source is searched in the data source mapping table according to the type of the business indicator to be queried (such as by analyzing the keywords or business classification identifiers in the business indicator). For example, if the business indicator to be queried is about the customer credit score, the server finds the corresponding credit score database in the data source mapping table as the data source for the query.

[0107] Alternatively, the data source can be determined by analyzing specific information in the business indicator to be queried. For example, if the business indicator to be queried contains account-related information (such as account number, account type, etc.), it can be determined that the query of the business indicator is related to the account database. If the business indicator involves multiple aspects (such as both customer information and transaction information), the main data source can be determined based on the priority set by the business rules or the purpose of the query. For example, if the purpose of the query is to understand the basic information of the customer when the customer conducts a certain transaction, the customer information database can be determined as the main data source.

[0108] S503: According to the database query instruction and the data source, query the database corresponding to the data source to obtain the query result.

[0109] The difference between the database corresponding to the data source and the aforementioned target database or website is that the target database or website is mainly used to display data that has been processed and summarized to facilitate staff viewing and interaction. The database corresponding to the data source stores more original and comprehensive data. If the RPA technology fails to obtain the expected matching result after querying the target database or website, it may be because the data on the target database or website has been filtered, aggregated, etc., resulting in the incomplete presentation of the required data or the inability to accurately match the query conditions in the website environment. For example, the target database or website may only display some common and macro data indicators, while the database corresponding to the data source contains more detailed indicator data for different business scenarios. For example, for customer label data, the target database or website may only display the summary information of some labels updated daily, while the database corresponding to the data source will have more label data with different update frequencies (such as monthly updates). Alternatively, the update mechanism of the data in the target database or website and the data corresponding to the data source may be different. The data update of the target database or website may be delayed or only update some important data, while the database corresponding to the data source can reflect the data changes more timely and completely. Even if the target database or website query does not obtain a matching result, the database corresponding to the data source may already have the latest data that meets the requirements. For example, for customer tag data with a timeliness of "updated daily", the database corresponding to the data source may contain new tag information after being updated on the same day, but the target database or website may not update and display this data until some time later.

[0110] Therefore, the query results can be obtained by querying the database corresponding to the data source according to the database query instruction and the data source, so as to obtain more accurate indicator query results.

[0111] The method provided in the embodiment of the present application generates a database query instruction for the business indicator to be queried according to the business indicator to be queried and a preset template if there is no first candidate data with a matching degree greater than or equal to a preset matching degree threshold in the query results of the target database or website. According to the business indicator to be queried, a data source for querying the business indicator to be queried through the database query instruction is determined. According to the database query instruction and the data source, a query result is obtained by querying in the database corresponding to the data source, thereby obtaining a more accurate indicator query result, thereby improving the accuracy of the indicator query.

[0112] Optional, based on the above Figure 5 Under the method shown, the embodiment of the present application may also include optimizing the content of the similarity model according to the new query results. Figure 6 A flowchart of another indicator query method provided in an embodiment of the present application is shown below. Figure 6As shown, the method may also include:

[0113] S601. Update the historical search data of the target database according to the query result.

[0114] Specifically, the query results can be parsed first to extract key information related to the historical search data. For example, if the query result is new credit rating information about a customer, the server will extract key information such as customer ID, credit rating value, query time, etc. Then, integrate these key information with the existing historical search data. If the historical search data is stored in the form of a database table, the new key information can be inserted into the corresponding table according to the data structure, or the relevant fields in the existing records can be updated. For example, if there is a table for storing customer credit-related query results in the historical search data, the new credit rating information will be updated to the corresponding customer record row; if it is a new customer, a new record containing customer ID, credit rating, query time and other information is inserted.

[0115] Optionally, during the integration process, data consistency checks and cleansing can be performed. For example, check whether newly inserted or updated data conforms to predefined data format requirements and remove possible duplicate or erroneous data. For example, if there are multiple credit rating query results for the same customer at the same query time (possibly due to data errors or repeated operations), they are processed according to business rules (such as taking the latest result) to ensure the accuracy and validity of historical search data.

[0116] In a possible implementation, the user's evaluation feedback on the query result may also be obtained, so as to update the historical search data according to the evaluation feedback and the query result.

[0117] In this implementation, the user's evaluation feedback on the query results can be obtained, and the evaluation feedback represents whether the query results meet the user's query expectations. According to the evaluation feedback and the query results, the historical search data is updated. Specifically, a special evaluation interaction area can be set in the interface that displays the query results to the user. For example, two buttons "satisfied" and "unsatisfied" are provided at the bottom of the query result page. When the user finishes viewing the query results, the user's operation in the evaluation interaction area is obtained. If the user clicks the "satisfied" button, a feedback signal representing a positive evaluation will be received; if the "unsatisfied" button is clicked, a negative evaluation feedback signal will be received. Alternatively, the user's evaluation feedback on the query results can also be obtained through satisfaction rating and other methods.

[0118] If the evaluation feedback is positive, the key information in the query result is marked. Then the record related to the query is searched in the historical search data. If there is a corresponding record, some related fields in the record are updated, such as raising the credibility mark of the query result (for example, the original credibility is 0.8, and it is raised to 0.9, indicating that the query result is more reliable). If there is no corresponding record, a new historical search data record is created based on the query result, and the credibility of the record is initialized to a higher value (such as 0.9).

[0119] If the evaluation feedback is negative, the key information in the query results will also be marked, and then relevant records will be found in the historical search data. For the relevant records found, their credibility marks will be lowered (for example, the original credibility was 0.8, and it was lowered to 0.6). If negative evaluation feedback for similar query results appears multiple times within a certain period of time (such as within a month), the server will mark the relevant query strategies or parameters in the historical search data for subsequent optimization and adjustment. For example, if the negative evaluation is caused by an inaccurate data source used when querying a certain type of business indicator, the server marks that the data source may need to be updated or the query method may need to be optimized.

[0120] S602: Optimize the similarity model according to the updated historical search data.

[0121] Use the updated historical search data to re-extract the data required for similarity model training. For example, according to the method previously used to train the similarity model, extract the first semantic feature vector of the historical search business indicator and the second semantic feature vector of the historical click result from the updated historical search data. Then, recalculate the similarity of the historical search business indicator and the historical click result based on the newly extracted data. For example, use the same cosine similarity algorithm or other similarity calculation method (such as the similarity of Euclidean distance transformation, etc.) to re-determine the similarity relationship between each pair of historical search business indicators and historical click results. Finally, use the recalculated first semantic feature vector, second semantic feature vector and similarity data to retrain the similarity model to optimize the similarity model.

[0122] The method provided in the embodiment of the present application updates the historical search data of the target database according to the query results, and optimizes the similarity model according to the updated historical search data, so that in the process of users continuously performing indicator queries, the similarity model is continuously optimized according to the generated search data, thereby improving the similarity judgment accuracy of the similarity model, and then improving the accuracy of the indicator query.

[0123] Figure 7 A flowchart of another indicator query method provided in an embodiment of the present application is shown below. Figure 7As shown, the method may also include:

[0124] S701. Store query results and attribute information of the query results.

[0125] The attribute information includes the validity range of the query result, which refers to the time interval during which the query result remains valid, and reflects how long the query result can accurately reflect the actual situation of the relevant business indicators.

[0126] Specifically, you can first create a dedicated data structure to store the query results and their attribute information. For example, you can use a database table structure containing multiple fields, one of which is used to store the query results (depending on the data type of the query results, it can be a value, a string, or a complex structured data), and the other fields are used to store the attribute information. When the query results are obtained, you can analyze the query results to determine their timeliness range. The query results and the determined attribute information (including the timeliness range) are stored in the created data structure in a predefined format.

[0127] S702: If a query request for a business indicator corresponding to a query result is received within a time limit, a query result of the query request is determined from the stored query results.

[0128] When a query request for a business indicator is received, the content of the business indicator to be queried in the query request can be first extracted. Then, before the query, a record matching the business indicator to be queried is searched from a data table, database, etc. storing the query results and the attribute information of the query results (for example, matching is performed by the name, number, etc. of the business indicator to be queried). During the search process, the time range field in the record is also checked.

[0129] If there is a matching record and the current time is within the time limit, the corresponding result can be directly extracted from the stored query results and returned to the requester as a response to this query request. If there is no matching record or the current time is not within the time limit, the above Figures 2 to 5 Any of the implementations described in the foregoing may be used to perform a conventional query process to obtain query results.

[0130] The method provided in the embodiment of the present application stores query results and attribute information of the query results including the time limit of the query results. If a query request for a business indicator corresponding to the query result is received within the time limit, the query result of the query request is determined from the stored query results. This method can determine whether there are query results within the time limit that can be directly obtained before performing RPA automatic query. If so, the query result is directly obtained, thereby further improving the efficiency of indicator query.

[0131] Figure 8 This is a schematic diagram of the structure of an indicator query device provided in an embodiment of the present application. Figure 8 As shown, the indicator query device may include: an acquisition module 11, a search module 12, a processing module 13, a control module 14, and an output module 15.

[0132] The acquisition module 11 is used to acquire the business indicator to be queried and the search priority of the business indicator, and the search priority is related to the attribute of the business indicator.

[0133] The search module 12 is used to search in the target database to obtain candidate data according to the business indicator to be queried and the search priority.

[0134] The processing module 13 is used to determine the matching degree between each candidate data and the business indicator to be queried according to a pre-trained similarity model, wherein the similarity model is generated based on the historical search records of the target database.

[0135] The control module 14 is configured to generate a query result of the business indicator to be queried according to the first candidate data if there is first candidate data with a matching degree greater than or equal to a preset matching degree threshold.

[0136] The output module 15 is used to output the query result.

[0137] Optionally, the processing module 13 is specifically used to input each candidate data and the business indicator to be queried into the similarity model, obtain a first matching score between each candidate data and the business indicator to be queried, and a second matching score between the attribute of each candidate data and the search priority. According to the first matching score and the second matching score, the matching degree of each candidate data and the business indicator to be queried is determined.

[0138] Optionally, when the search priority includes at least two priorities, the search module 12 is specifically used to search the target database according to the attributes of the business indicator corresponding to the first priority and the business indicator to be queried to obtain the search result. If the search result is empty, the target database is searched according to the attributes of the business indicator corresponding to the second priority and the business indicator to be queried to obtain the search result, and the second priority is located at the next priority after the first priority. If the search result is not empty, the search result is used as the candidate data.

[0139] Optionally, the similarity model is generated based on the training of historical search records of the target database, including: obtaining the historical search records of the target database, the historical search records including historical search business indicators and historical click results. Extracting the first semantic feature vector of the historical search business indicator and the second semantic feature vector of the historical click results based on the historical search records. Determine the similarity between the historical search business indicator and the historical click result based on the first semantic feature vector and the second semantic feature vector. Train an initial model based on the first semantic feature vector, the second semantic feature vector, and the similarity between the historical search business indicator and the historical click result to generate the similarity model.

[0140] Optionally, if there is no first candidate data with a matching degree greater than or equal to a preset matching degree threshold, the control module 14 is further used to generate a database query instruction for the business indicator to be queried according to the business indicator to be queried and a preset template. The processing module 13 is further used to determine, according to the business indicator to be queried, a data source for querying the business indicator to be queried through the database query instruction. The search module 12 is further used to query the database corresponding to the data source according to the database query instruction and the data source to obtain the query result.

[0141] Optionally, the processing module 13 is further configured to update the historical search data of the target database according to the query result, and optimize the similarity model according to the updated historical search data.

[0142] Optionally, the acquisition module 11 is further used to acquire the user's evaluation feedback on the query result, the evaluation feedback indicating whether the query result meets the user's query expectation. The processing module 13 is further used to update the historical search data according to the evaluation feedback and the query result.

[0143] Optionally, the control module 14 is further configured to store the query result and the attribute information of the query result, wherein the attribute information includes the time limit of the query result. If a query request for a business indicator corresponding to the query result is received within the time limit, the query result of the query request is determined from the stored query result.

[0144] The indicator query device provided in the embodiment of the present application can execute the indicator query method in the above method embodiment. Its implementation principle and technical effect are similar and will not be repeated here.

[0145] Fig. 9 The following is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device is used to execute the aforementioned index query method, for example, it can be the aforementioned electronic device with data processing function. Fig. 9As shown, the electronic device 900 may include: at least one processor 901 , a memory 902 , and a communication interface 903 .

[0146] The memory 902 is used to store programs. Specifically, the programs may include program codes, and the program codes include computer operation instructions.

[0147] The memory 902 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0148] The processor 901 is used to execute the computer-executable instructions stored in the memory 902 to implement the method described in the above method embodiment. The processor 901 may be a CPU, or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0149] The processor 901 can communicate and interact with external devices through the communication interface 903. The external device can be, for example, the target database mentioned above. In specific implementation, if the communication interface 903, the memory 902 and the processor 901 are implemented independently, the communication interface 903, the memory 902 and the processor 901 can be connected to each other through a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, etc., but it does not mean that there is only one bus or one type of bus.

[0150] Optionally, in a specific implementation, if the communication interface 903, the memory 902 and the processor 901 are integrated on a chip, the communication interface 903, the memory 902 and the processor 901 can communicate through an internal interface.

[0151] The present application also provides a computer-readable storage medium, which may include: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, and other media that can store program codes. Specifically, the computer-readable storage medium stores program instructions, and the program instructions are used for the methods in the above embodiments.

[0152] The present application also provides a program product, which includes an execution instruction, which is stored in a readable storage medium. At least one processor of a computing device can read the execution instruction from the readable storage medium, and at least one processor executes the execution instruction so that the computing device implements the above-mentioned indicator query method.

[0153] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, 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 optional embodiments, and the actions and modules involved are not necessarily required by the present application.

[0154] It should be further noted that, although the various steps in the flowchart are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowchart may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these sub-steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0155] It should be understood that the above-mentioned device embodiments are only illustrative, and the device of the present application can also be implemented in other ways. For example, the division of units / modules in the above-mentioned embodiments is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units, modules or components can be combined, or can be integrated into another system, or some features can be ignored or not executed.

[0156] In addition, unless otherwise specified, each functional unit / module in each embodiment of the present application may be integrated into one unit / module, each unit / module may exist physically separately, or two or more units / modules may be integrated together. The above-mentioned integrated unit / module may be implemented in the form of hardware or in the form of a software program module.

[0157] If the integrated unit / module is implemented in the form of hardware, the hardware may be a digital circuit, an analog circuit, etc. The physical implementation of the hardware structure includes but is not limited to transistors, memristors, etc. Unless otherwise specified, the processor may be any appropriate hardware processor, such as a CPU, a GPU, an FPGA, a DSP, an ASIC, etc. Unless otherwise specified, the storage unit may be any appropriate magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory (RRAM), a dynamic random access memory (DRAM), a static random access memory (SRAM), an enhanced dynamic random access memory (EDRAM), a high-bandwidth memory (HBM), a hybrid memory cube (HMC), etc.

[0158] If the integrated unit / module is implemented in the form of a software program module 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. The computer software product is stored in a memory, including a number of instructions to enable a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned memory includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or CD-ROM and other media that can store program codes.

[0159] In the above embodiments, the description of each embodiment has its own emphasis. For the part not described in detail in a certain embodiment, please refer to the relevant description of other embodiments. The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0160] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present application. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0161] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. An indicator query method, characterized in that: include: Acquire a business indicator to be queried, and a search priority of the business indicator, where the search priority is related to an attribute of the business indicator; Searching in a target database to obtain candidate data according to the business indicator to be queried and the search priority; Determining the matching degree between each candidate data and the business indicator to be queried according to a pre-trained similarity model, wherein the similarity model is generated and trained based on the historical search records of the target database; If there is first candidate data whose matching degree is greater than or equal to the preset matching degree threshold, generating a query result of the business indicator to be queried according to the first candidate data; The query result is output.

2. The method according to claim 1, characterized in that: Determining the matching degree between each candidate data and the business indicator to be queried according to the pre-trained similarity model includes: Inputting each of the candidate data and the business indicator to be queried into the similarity model to obtain a first matching score between each of the candidate data and the business indicator to be queried, and a second matching score between an attribute of each of the candidate data and the search priority; The matching degree between each of the candidate data and the business indicator to be queried is determined according to the first matching score and the second matching score.

3. The method according to claim 2, characterized in that The search priority includes at least two priorities, and searching in the target database to obtain candidate data according to the business indicator to be queried and the search priority includes: According to the attributes of the business indicator corresponding to the first priority and the business indicator to be queried, searching in the target database to obtain search results; If the search result is empty, searching the target database to obtain search results according to the attributes of the business indicator corresponding to the second priority and the business indicator to be queried, the second priority being one priority level after the first priority level; If the search result is not empty, the search result is used as the candidate data.

4. The method according to claim 1, characterized in that The similarity model is generated based on the historical search records of the target database, including: Acquire historical search records of the target database, wherein the historical search records include historical search service indicators and historical click results; Extracting a first semantic feature vector of the historical search service indicator and a second semantic feature vector of the historical click result according to the historical search record; Determining the similarity between the historical search service indicator and the historical click result according to the first semantic feature vector and the second semantic feature vector; An initial model is trained according to the first semantic feature vector, the second semantic feature vector, and the similarity between the historical search service indicator and the historical click result to generate the similarity model.

5. The method according to any one of claims 1 to 4, characterized in that If there is no first candidate data with a matching degree greater than or equal to a preset matching degree threshold, the method further includes: Generate a database query instruction for the business indicator to be queried according to the business indicator to be queried and a preset template; According to the business indicator to be queried, determining a data source for querying the business indicator to be queried through the database query instruction; According to the database query instruction and the data source, the query result is obtained by querying in the database corresponding to the data source.

6. The method according to claim 5, characterized in that The method further comprises: According to the query result, updating the historical search data of the target database; The similarity model is optimized according to the updated historical search data.

7. The method according to claim 6, characterized in that The updating of the historical search data of the target database according to the query result includes: Obtaining user evaluation feedback on the query result, wherein the evaluation feedback indicates whether the query result meets the user's query expectation; The historical search data is updated according to the evaluation feedback and the query result.

8. The method according to claim 5, characterized in that The method further comprises: storing the query result and attribute information of the query result, wherein the attribute information includes a time validity range of the query result; If a query request for a business indicator corresponding to the query result is received within the time limit, the query result of the query request is determined from the stored query results.

9. An indicator query device, characterized in that: include: An acquisition module, used to acquire the business indicator to be queried, and a search priority of the business indicator, wherein the search priority is related to an attribute of the business indicator; A search module, used to search in a target database to obtain candidate data according to the business indicator to be queried and the search priority; A processing module, used to determine the matching degree between each candidate data and the business indicator to be queried according to a pre-trained similarity model, wherein the similarity model is generated based on the training of the historical search records of the target database; A control module, configured to generate a query result of the business indicator to be queried according to the first candidate data if there is first candidate data with a matching degree greater than or equal to a preset matching degree threshold; An output module is used to output the query result.

10. An electronic device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 8 when executed by a processor.

12. A computer program product, characterized in that The invention comprises a computer program, which implements the method according to any one of claims 1 to 8 when being executed by a processor.

Images