Data retrieval method, apparatus, device, and storage medium

Abstract

Embodiments of the present application provide a data retrieval method, device and equipment and storage medium, relating to the technical field of data analysis. The method comprises: in response to a retrieved request, determining a target search term and a request object; based on an object search term library corresponding to the request object and a pre-constructed business system search term library, performing word segmentation processing on the target search term respectively, and based on the word segmentation processing result corresponding to the business system search term library and a pre-set priority label, establishing a corresponding search index; and based on the business system corresponding to the request object, the word segmentation result obtained based on the object search term library and the search index, obtaining a search result. The embodiments of the present application solve the problem that the search result cannot meet the user's expectation when the search engine searches diversified data in the prior art.

Description

Technical Field

[0001] This application relates to the field of data analysis technology, and in particular to a data retrieval method, apparatus, device, and storage medium. Background Technology

[0002] Most existing enterprise-built search engines are constructed using traditional retrieval methods. Their functional code is integrated into the enterprise's business system itself, and the data sources of their index files are usually limited to data within the business system. They adjust the hit ranking of search results by building a dedicated business thesaurus and allocating computational weights to relevant domains. When the data to be searched is not limited to the business system but includes diversified data including external data, it is difficult to guarantee that the search results will meet the user's expectations. Summary of the Invention

[0003] This application provides a data retrieval method, apparatus, device, and storage medium to address the problem in the prior art where search results fail to meet user expectations when searching for diverse data.

[0004] In a first aspect, embodiments of this application provide a data retrieval method, which includes:

[0005] In response to the received search request, determine the corresponding target search terms and the request object;

[0006] Based on the object retrieval thesaurus corresponding to the request object and the pre-built business system retrieval thesaurus, the target retrieval terms are segmented. The object retrieval thesaurus corresponding to the request object is built based on the retrieval history of the request object, and the business system retrieval thesaurus is built based on data from the local business system and data from external databases.

[0007] Based on the word segmentation results and pre-set priority tags of the business system's retrieval thesaurus, a corresponding retrieval index is established;

[0008] The search results are obtained based on the business system corresponding to the request object, the word segmentation results obtained from the object search thesaurus, and the search index.

[0009] Optionally, the business system retrieval thesaurus is obtained through the following methods: acquiring local business system data and external data collected from external databases; obtaining corresponding structured data copies based on the structured data in the local business system data and external data; obtaining corresponding entity graphs based on the unstructured data in the local business system data and external data, the entity graphs being used to represent the relationships between the local business system data and external data; adding corresponding source tags to the structured and unstructured data respectively based on the specific sources of the local business system data and external data; adding format tags to the unstructured data based on the file format of the unstructured data; and obtaining the business system retrieval thesaurus based on the structured data copies, entity graphs, source tags, and format tags.

[0010] Optionally, based on the structured data in the local business system data and the external data, a corresponding structured data copy is obtained, including: establishing a primary key association between the structured data and the database corresponding to the local business system data; integrating the external data based on the primary key association; and combining the integration result with the local business system data based on the primary key association to generate a structured data copy.

[0011] Optionally, based on unstructured data from local business system data and external data, a corresponding entity graph is obtained, including: extracting target information from the unstructured data, the target information including entities, entity attributes, and classification labels; based on the target information, obtaining the matching degree between unstructured data through correlation analysis; extracting the corresponding terms of entities in the target information and marking the matching degree on the terms; based on the matching degree, establishing a graph database corresponding to the entities in the target information and generating an entity graph.

[0012] Optionally, based on the specific sources of local business system data and external data, corresponding source tags are added to structured data and unstructured data respectively, including: adding source labels corresponding to structured data in structured data copies; setting corresponding priority tags according to data sources in sub-data within the same structured data copy; and adding source tags corresponding to each entity in the graph database of unstructured data.

[0013] Optionally, external data is collected in the following ways: if the external data is structured data, monitor the data changes in the external database of the corresponding external business system in real time; collect the structured data in the external data according to the type of data content corresponding to the data changes; if the external data is unstructured data, obtain the push request from the external business system; and obtain the external data corresponding to the external business system based on the push request.

[0014] Optionally, depending on the type of data content corresponding to the data change, structured data from external data is collected, including: if the data content is a single table data change, a pull task is sent to the external business system to extract the single table data information; if the data content is continuously incremental change information, a subscription task is sent to the external business system to continuously obtain real-time incremental change information through subscription; if the data content is existing data that has not been collected, all external data information is pulled.

[0015] Optionally, based on the push request, external data is obtained, including: collecting unstructured data from the external data to a local storage location based on a set transmission method; and extracting structured data information from the unstructured data in the local storage location through a set collection method.

[0016] Optionally, after obtaining external data based on a push request, the method further includes: verifying the legality of the obtained external data; de-identifying the business-sensitive data in the external data that has passed the legality verification; and saving the external data to the corresponding database shard or table based on the external business system corresponding to the de-identified external data.

[0017] Optionally, based on the word segmentation processing results corresponding to the business system's retrieval thesaurus and the pre-set priority tags, a corresponding retrieval index is established, including: performing an inverted index on the inventory data based on the word segmentation processing results corresponding to the business system's retrieval thesaurus; obtaining the word segmentation domain priority based on the priority tags contained in the structured data copies in the inventory data and the word segmentation processing results corresponding to the business system's retrieval thesaurus; and establishing a retrieval index based on the inverted index and the word segmentation domain priority.

[0018] Optionally, based on the business system corresponding to the requested object, the word segmentation results obtained from the object retrieval thesaurus, and the retrieval index, the retrieval results are obtained, including: determining the retrieval preferences of the requested object based on the word segmentation results obtained from the object retrieval thesaurus; determining the data in the business system corresponding to the object as the main retrieval data; determining the data in external business systems associated with the business system corresponding to the object as the relevant retrieval data based on the retrieval preferences and entity graph; and retrieving the main retrieval data and the relevant retrieval data based on the retrieval index, and displaying the retrieved data corresponding to the main retrieval data and the relevant retrieval data respectively to obtain the retrieval results.

[0019] Optionally, based on the business system corresponding to the requested object, the word segmentation results obtained from the object search thesaurus, and the search index, the search results are obtained, including: obtaining user-defined search configuration; and searching local business system data and external data based on the search configuration and the search index to obtain search results.

[0020] Secondly, embodiments of this application provide a data retrieval device, which includes:

[0021] The acquisition module is used to determine the corresponding target search terms and request objects in response to the acquired search request;

[0022] The word segmentation module is used to segment target search terms based on the search term library corresponding to the request object and the pre-built business system search term library. The search term library corresponding to the request object is built based on the search history of the request object, and the business system search term library is built based on data from the local business system and data from external databases.

[0023] The index module is used to build a corresponding search index based on the word segmentation results and pre-set priority tags of the business system's retrieval thesaurus.

[0024] The output module is used to build a corresponding search index based on the word segmentation results of the business system's search thesaurus and the pre-set priority tags.

[0025] Optionally, the word segmentation module includes obtaining the business system retrieval lexicon through the following methods: acquiring local business system data and external data collected from external databases; obtaining corresponding structured data copies based on the structured data in the local business system data and external data; obtaining corresponding entity graphs based on the unstructured data in the local business system data and external data, the entity graphs being used to represent the relationships between the local business system data and external data; adding corresponding source tags to the structured data and unstructured data respectively based on the specific sources of the local business system data and external data; adding format tags to the unstructured data based on the file format of the unstructured data; and obtaining the business system retrieval lexicon based on the structured data copies, entity graphs, source tags, and format tags.

[0026] Optionally, the word segmentation module is specifically used to: establish a primary key association between the structured data and the local business system data in the database; integrate external data based on the primary key association; and combine the integration result with the local business system data based on the primary key association to generate a copy of the structured data.

[0027] Optionally, the word segmentation module is specifically used to: extract target information from unstructured data, including entities, entity attributes, and classification labels; based on the target information, obtain the matching degree between unstructured data through correlation analysis; extract the corresponding terms of entities in the target information and mark the matching degree on the terms; based on the matching degree, establish a graph database corresponding to the entities in the target information and generate an entity graph.

[0028] Optionally, the word segmentation module is specifically used to: add source annotations corresponding to the structured data in the structured data copy; set corresponding priority tags according to the data source in the sub-data in the same structured data copy; and add source tags corresponding to each entity in the graph database of unstructured data.

[0029] Optionally, the word segmentation module includes collecting external data in the following ways: if the external data is structured data, monitor the data changes in the external database of the corresponding external business system in real time; collect structured data in the external data according to the type of data content corresponding to the data changes; if the external data is unstructured data, obtain the push request from the external business system; and obtain the external data corresponding to the external business system based on the push request.

[0030] Optionally, the word segmentation module is specifically used to: if the data content is a single table data change, send a pull task to the external business system to extract the single table data information; if the data content is continuously incremental change information, send a subscription task to the external business system to continuously obtain real-time incremental change information through subscription; if the data content is existing data that has not been collected, pull all external data information.

[0031] Optionally, the word segmentation module is specifically used to collect unstructured data from external data to a local storage location by setting a transmission method; and to extract structured data information from the unstructured data in the local storage location by setting a collection method.

[0032] Optionally, the word segmentation module is also used to: after obtaining external data based on a push request, perform legality verification on the obtained external data; perform desensitization processing on the business-sensitive data in the external data that has passed the legality verification; and save the external data to the corresponding database shard or table based on the external business system corresponding to the desensitized external data.

[0033] Optionally, the indexing module is specifically used to: perform an inverted index on the inventory data based on the word segmentation processing results corresponding to the business system's retrieval thesaurus; obtain the word segmentation domain priority based on the priority tags contained in the structured data copies in the inventory data and the word segmentation processing results corresponding to the business system's retrieval thesaurus; and establish a retrieval index based on the inverted index and the word segmentation domain priority.

[0034] Optionally, the output module is specifically used to: determine the search preferences of the requested object based on the word segmentation results obtained from the object search thesaurus; determine the data in the business system corresponding to the object as the main search data; determine the data in the external business system associated with the business system corresponding to the object as the relevant search data based on the search preferences and entity graph; and search the main search data and relevant search data based on the search index, and display the search results for the main search data and relevant search data respectively.

[0035] Optionally, the output module is specifically used to obtain user-defined search configurations; based on the search configurations and search indexes, to search local business system data and external data to obtain search results.

[0036] Thirdly, embodiments of this application also provide a control device, which includes:

[0037] At least one processor;

[0038] and memory that is communicatively connected to at least one processor;

[0039] The memory stores instructions that can be executed by at least one processor, which, when executed by at least one processor, cause the control device to perform a data retrieval method corresponding to any embodiment of the first aspect of the present application.

[0040] Fourthly, embodiments of this application also provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement any of the data retrieval methods described in the first aspect of embodiments of this application.

[0041] Fifthly, embodiments of this application also provide a computer program product, which includes computer execution instructions that, when executed by a processor, are used to implement any of the data retrieval methods described in the first aspect of embodiments of this application.

[0042] The data retrieval method, apparatus, device, and storage medium provided in this application, in response to a received retrieval request, determine the corresponding search terms and the requested object; based on the search term library corresponding to the requested object and a pre-built business system search term library, perform word segmentation processing on the target search terms respectively; based on the word segmentation processing results corresponding to the business system search term library and pre-set priority tags, establish a corresponding search index; and obtain the retrieval results based on the business system corresponding to the requested object, the word segmentation results obtained based on the object search term library, and the search index. Therefore, the obtained retrieval results can simultaneously integrate proprietary data from the local business system and data from external databases, improving the retrieval efficiency and accuracy of data from external databases, thereby improving the accuracy of the retrieval results. Attached Figure Description

[0043] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0044] Figure 1 An application scenario diagram of the data retrieval method provided in the embodiments of this application;

[0045] Figure 2 A flowchart illustrating a data retrieval method provided in one embodiment of this application;

[0046] Figure 3a A flowchart illustrating a data retrieval method provided in yet another embodiment of this application;

[0047] Figure 3b for Figure 3a Flowchart of the method for collecting exogenous data in the illustrated embodiment;

[0048] Figure 3c for Figure 3a The flowchart of the method for generating structured data copies in the illustrated embodiment is shown.

[0049] Figure 3d for Figure 3a The flowchart of the method for obtaining entity maps corresponding to unstructured data in the embodiment shown is as follows;

[0050] Figure 3e for Figure 3a The flowchart of the method for adding data source tags in the illustrated embodiment is shown.

[0051] Figure 4 This is a schematic diagram of the structure of a data retrieval device provided in yet another embodiment of this application;

[0052] Figure 5 This is a schematic diagram of the structure of a control device provided in yet another embodiment of this application.

[0053] The accompanying drawings have illustrated specific embodiments of this disclosure, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concepts of this disclosure to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0054] In the following description, when referring to the accompanying drawings, the same numbers in different drawings denote the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this application as detailed in the appended claims.

[0055] The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0056] With the advent of the big data era, business information (such as drug information in pharmaceutical companies and material procurement information in construction companies) is showing a trend of diversification and multi-level volume, and users' demand for information retrieval is also increasing. Faced with the diversification of data content, search engines built on traditional retrieval technologies suffer from information overload, returning too many search results, and having a single format of search results. They are unable to retrieve all kinds of related data with multiple structures in vertical fields, resulting in a high frequency of secondary searches by users. Furthermore, as the amount of indexed data increases, search engines built on traditional retrieval technologies inevitably experience the problem of extended retrieval time, thereby increasing the response time to requests and affecting the user experience.

[0057] Most existing enterprise-built search engines are constructed using traditional retrieval methods. Their functional code is integrated into the enterprise's business system, and the data sources for their index files are usually limited to data within the business system. They adjust the ranking of search results by building a dedicated business thesaurus and allocating computational weights to relevant domains. When the data to be searched is not limited to the business system but includes diverse data from external sources, they can usually only use batch data push methods (such as displaying the number of related data entries without displaying the specific content). It is difficult to effectively classify and analyze the data. Furthermore, because they cannot collect and store external data in real time, search engines built using existing technologies have a T+1 time limit when searching for diverse data.

[0058] To address this issue, this disclosure provides a data retrieval method that pre-establishes a business system retrieval thesaurus containing data from external databases and local business systems. This allows the retrieval results to cover data from both the local business systems and external databases, effectively ensuring that the retrieval results simultaneously integrate data from both systems and improve the accuracy of the results.

[0059] The application scenarios of the embodiments of this disclosure are explained below:

[0060] Figure 1 This diagram illustrates an application scenario of the data retrieval method provided in this embodiment of the disclosure. For example... Figure 1As shown, the retrieval system 100 collects data from the local business system 110 and the external business system 120 in real time, integrates the collected data, and when it receives a retrieval request from the requesting object 130, it outputs the corresponding retrieval results to the requesting object 130 through aggregate retrieval, thus completing the data retrieval.

[0061] It should be noted that, Figure 1 In the scenario shown, only one of the retrieval system, local business system, external business system, and request object is used for illustration. However, this application embodiment is not limited to this. That is to say, the number of retrieval system, local business system, external business system, and request object can be arbitrary.

[0062] The data retrieval method provided in this application is described in detail below through specific embodiments. It should be noted that the following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0063] Figure 2 A flowchart illustrating a data retrieval method provided in one embodiment of this disclosure. Figure 2 As shown, the data retrieval method provided in this embodiment includes the following steps:

[0064] Step S201: In response to the obtained search request, determine the corresponding target search term and request object.

[0065] Specifically, the request recipient is the user who issued the search request. This user is typically a user of the business system, not a user of a conventional search engine (the conventional search engine used by these users is generally not associated with the business system). Depending on the user, the search system will store different user search histories and the business system to which the user belongs, all of which will affect the search results. Therefore, when determining a search request, the request recipient will be identified, and further information such as their search history and the corresponding business system will be determined.

[0066] Step S202: Based on the search term library corresponding to the request object and the pre-built business system search term library, perform word segmentation on the target search terms respectively.

[0067] The search term library corresponding to the request object is built based on the search history of the request object, while the search term library of the business system is built based on data from the local business system and data from external databases.

[0068] Specifically, the search history of the requested object is stored in the search system, thereby enabling the creation of a search term database for the user's corresponding object.

[0069] By segmenting search terms using an object search thesaurus, we can quickly identify search targets that users may be interested in, and then accurately push search results.

[0070] The business system's search term library includes data from both the local business system and external databases. Corresponding tags are added to entities in both types of data. For example, in a warehousing business system, goods recorded in inventory by the local business system are entities, and goods that can be added to inventory by external databases are also entities. The type, quantity, and location of these goods are their corresponding tags or related information, which are the search terms in the search term library.

[0071] By segmenting the target search terms using the business system's thesaurus, it is possible to quickly determine the entities and related information in the business system that correspond to the target search terms, and thus determine the search results that can be pushed to users.

[0072] By combining object-oriented search terminology and business system search terminology, data that can be pushed to users is determined from two perspectives: user preferences and system inventory data. This approach covers data from both local business systems and external databases, thereby improving the accuracy of search results.

[0073] Step S203: Based on the word segmentation results and pre-set priority tags of the business system's retrieval thesaurus, establish the corresponding retrieval index.

[0074] Specifically, the preset priority tags are mainly used to distinguish the priorities among data that can be pushed, as determined by word segmentation. For example, in the pharmaceutical field, when searching for a certain drug, it may simultaneously retrieve commonly used varieties with large stock in pharmacies, less commonly used varieties related to it, and varieties that can be replenished by external companies (i.e., data from external databases). By assigning different priorities to the source categories to which they belong, their corresponding priority tags can be obtained.

[0075] Based on priority tags and word segmentation results, the data corresponding to the target search term in the business system, and its push priority, can be determined. This determined data and its priority constitute the search index. Using the search index, corresponding data can be pushed as search results according to priority.

[0076] Step S204: Based on the business system corresponding to the request object, the word segmentation results obtained from the object retrieval thesaurus, and the retrieval index, obtain the retrieval results.

[0077] Specifically, when determining search results, in addition to relying on the search index, it is also necessary to consider the business system corresponding to the requested object. For example, different departments of a scientific research enterprise may be involved in research and development work in different fields. Therefore, when pushing search results to them, data in the same field (i.e., the same business system) in the search index will be given priority as the pushed data. Similarly, the data corresponding to the word segmentation results obtained from the object search thesaurus in the search index will also be given priority as the pushed results. In this way, the search results can be combined with user behavior, making the data pushed to users more accurate.

[0078] The data retrieval method provided in this disclosure, in response to a received retrieval request, determines the corresponding search terms and the requested object; based on the retrieval terminology corresponding to the requested object and a pre-built business system retrieval terminology, it performs word segmentation on the target search terms respectively; based on the word segmentation results corresponding to the business system retrieval terminology and pre-set priority tags, it establishes a corresponding retrieval index; and based on the business system corresponding to the requested object, the word segmentation results obtained from the object retrieval terminology, and the retrieval index, it obtains the retrieval results. Therefore, the obtained retrieval results can simultaneously integrate proprietary data from the local business system and data from external databases, improving the retrieval efficiency and accuracy of data from external databases, thereby improving the accuracy of the retrieval results.

[0079] Figure 3a A flowchart illustrating a data retrieval method provided in one embodiment of this disclosure. Figure 3a As shown, the data retrieval method provided in this embodiment includes the following steps:

[0080] Step S301: Obtain local business system data and external data collected from external databases.

[0081] Specifically, the retrieval system acquires data from the local business system in real time and also collects external data from external databases. Alternatively, it can collect and store both types of data through a dedicated data acquisition module or platform (this article will describe the implementation of the data acquisition part using the data acquisition platform as an example). Real-time acquisition effectively ensures the timeliness of the acquired external data, improving data timeliness and availability (since data from the local business system can be directly obtained, its timeliness is inherently guaranteed).

[0082] Furthermore, such as Figure 3b The diagram shows a flowchart of the method for collecting exogenous data. The exogenous data is collected in the following way:

[0083] Step S3011: If the external data is structured data, monitor the data changes in the external database of the corresponding external business system in real time.

[0084] Specifically, external data includes structured and unstructured data. Structured data includes forms and text updated on fixed pages or at fixed addresses. Unstructured data includes downloaded attachments corresponding to randomly generated addresses provided by external databases. Unstructured data typically has random addresses, formats, and sizes, making it impossible to quickly and in batches extract its content using structured methods and import it into the data acquisition platform.

[0085] Furthermore, the data acquisition platform does not continuously collect data from every external database in real time (e.g., once per second). This is because some external databases may not have been updated for a period of time. Therefore, the data acquisition platform typically determines whether to collect data from external databases by monitoring changes in the data. If changes occur in an external database, only the changed portion of the external data can be collected, thereby reducing the amount of data to be collected.

[0086] In some embodiments, the data acquisition platform establishes associations with external business systems that require data collection to determine which external database data corresponding to which external business systems need to be collected. For example, a data acquisition platform for the pharmaceutical industry only needs to collect external database data corresponding to pharmaceutical external business systems, and does not need to collect data corresponding to external business systems in other fields (such as the decoration industry).

[0087] In some embodiments, the data acquisition platform sends monitoring tasks to the external databases corresponding to the external business systems with which it has established relationships, so as to monitor data changes in the database in real time through the monitoring tasks.

[0088] Step S3012: Collect structured data from external data according to the type of data content corresponding to the data changes.

[0089] Specifically, the type of data portion (i.e., data content) that changes in the external database will affect the specific collection method and the specific data content collected, therefore, they need to be discussed separately.

[0090] Furthermore, the methods for collecting data according to different data content types specifically include:

[0091] Scenario 1 (not shown): If the data content is a single table data change, send a pull task to the external business system to extract the single table data information.

[0092] Specifically, if the data content is data from a certain form (i.e., a single table), then you can directly extract the data information from that form.

[0093] Scenario 2 (not shown): If the data content is continuously incrementally changing information, send a subscription task to the external business system to continuously obtain real-time incremental change information through subscription.

[0094] Specifically, if the data content is information that is continuously changing in the same address or in addresses with a fixed pattern, or information that is continuously updated by the same publisher, then you can directly subscribe to or obtain the updated content (i.e., incremental change information).

[0095] Case 3 (not shown): If the data content is existing data that has not been collected, retrieve all external data information.

[0096] Specifically, if the data content itself is content that the data acquisition platform has not collected before, there is no need to wait for it to update again or confirm which parts of the data have changed; instead, all data content can be obtained directly. For example, if a department has newly established a device system for its corresponding business system, and all its data is newly configured, then the configuration data corresponding to that device system can be obtained directly.

[0097] Step S3013: If the external data is unstructured data, obtain the push request from the external business system.

[0098] Specifically, since unstructured data cannot be retrieved as quickly as structured data, its corresponding address is usually obtained based on push requests from external business systems, rather than being directly fetched.

[0099] Step S3014: Based on the push request, obtain the external data corresponding to the external business system.

[0100] Specifically, since the address of external data cannot be predetermined and directly retrieved, it is necessary to obtain the corresponding data based on the push request after receiving the push request.

[0101] Furthermore, since exogenous data is unstructured, its specific acquisition methods include:

[0102] Step 1 (not shown): Collect unstructured data from external data to local storage location based on the set transmission method.

[0103] Specifically, unstructured data can be collected via FTP (File Transfer Protocol).

[0104] Step 2 (not shown): Extract structured data information from unstructured data in local storage by setting the acquisition method.

[0105] Specifically, since unstructured data may have different formats, the specific collection methods can be POI (Apache POI, an open-source package for extracting text from unstructured documents), ICR (Intelligent Character Recognition), OCR (optical character recognition), etc.

[0106] Step S3015: Perform a legality check on the obtained external data.

[0107] Specifically, after obtaining external data, it is necessary to perform a legality verification to determine the legality of the data.

[0108] Step S3016: De-identify sensitive business data in external data that has passed the legality verification.

[0109] Specifically, since the data in the business system may contain sensitive data, such as specific parameters involving confidentiality requirements, it is necessary to anonymize this sensitive data based on the configured confidentiality requirements.

[0110] Step S3017: Based on the external business system corresponding to the de-identified external data, save the external data to the corresponding database shard or table.

[0111] Specifically, data acquisition platforms typically store the collected external data in a database, and based on the external business system from which the external data originates, they divide the data into separate databases or tables for easy retrieval.

[0112] Step S302: Based on the structured data in the local business system data and the external data, obtain the corresponding structured data copy.

[0113] Specifically, after collecting local business system data and external data, it is necessary to further integrate the data based on whether it is structured data, in order to improve data correlation, thereby increasing the degree of aggregation during retrieval and enhancing the reliability of retrieval results.

[0114] In some embodiments, the data integration function in the retrieval system can be accomplished by a dedicated integration module or a data aggregation platform (in this embodiment, the subsequent description of the integration part will use the data aggregation platform instead of the retrieval system).

[0115] Furthermore, such as Figure 3c The diagram shown is a flowchart of a method for generating structured data copies. This method specifically includes:

[0116] Step S3021: Establish primary key associations between the structured data and the local business system data in the database.

[0117] Specifically, since both structured data in the local business system and structured data from external sources are stored in the database, and due to their structured nature, they can be quickly located using the primary key value, the first step is to establish the association between the primary keys of the structured data and the data in the local business system. This means determining the correspondence between the structured data in the local business system and the primary keys in the corresponding database of the local business system, such as which primary key a certain piece of data has and what its key value is. Through this correspondence, the corresponding data can be quickly identified using the primary key and its key value.

[0118] Step S3022: Integrate external data based on primary key association.

[0119] Specifically, after establishing the corresponding primary key relationships for structured data in the local business system, it is also necessary to establish the corresponding primary key relationships for the external data (structured data in the data acquisition platform's database). The method is similar to the integration of structured data in the local business system, and will not be elaborated here.

[0120] Step S3023: Combine the integration results with the local business system data based on the primary key association to generate a structured data copy.

[0121] Specifically, a structured data copy is a separate file that stores the mapping between structured data in the local business system and structured data from external sources, and the primary keys of the database in the data acquisition platform. This copy file allows for the rapid determination of the location of the corresponding structured data, and then the reading of that structured data.

[0122] Step S303: Based on the unstructured data in the local business system data and the external data, obtain the corresponding entity map.

[0123] Among them, the entity graph is used to represent the relationship between local business system data and external data.

[0124] Specifically, since unstructured data cannot usually be linked through primary keys (because the data format, size, source, etc. are different, the storage path and location are usually not standardized), a graph database can be used to establish relationships between data and obtain an entity graph of unstructured data based on these relationships. Similar to structured copies, the entity graph can quickly locate the position of unstructured data and then read it.

[0125] Furthermore, such as Figure 3dThe diagram shows a flowchart of a method for obtaining entity maps corresponding to unstructured data. The specific methods for obtaining entity maps include:

[0126] Step S3031: Extract target information from unstructured data.

[0127] The target information includes entities, entity attributes, and category tags.

[0128] Specifically, unstructured data generally contains attribute information or configuration information that can indicate its corresponding entity, such as the file name and format extension of a compressed file, or the specific attributes of an audio file. Among this information, the part containing the entity, entity attributes and its corresponding category label is the target information needed for retrieval.

[0129] The category tags can be the business domain to which the entity belongs, or specific technology or format categories. For example, an entity may contain both "audio" and "orthopedic surgery," corresponding to format and technology categories, respectively. An entity can contain multiple category tags to facilitate retrieval.

[0130] Step S3032: Based on the target information, obtain the matching degree between unstructured data through correlation analysis.

[0131] Specifically, after obtaining the target information of unstructured data, the matching degree between each entity can be calculated. For example, if two entities are files of the same format and in the same field, their matching degree can be high, while files of different fields and different formats may have a lower matching degree.

[0132] The specific method for calculating the matching degree can be any existing correlation analysis method, and there are no restrictions here.

[0133] Step S3033: Extract the corresponding terms from the target information and mark the matching degree on the terms.

[0134] Specifically, entity names can be very long, containing either information needed for retrieval or information not needed, such as garbled characters or formatting symbols. Therefore, extracting the necessary information from the entity yields the corresponding term. By annotating the terms with the matching degree between entities calculated in the preceding steps, the matching degree between terms is obtained. Based on this matching degree, the matching degree between corresponding entities can be directly determined.

[0135] Step S3034: Based on the matching degree, establish a graph database corresponding to the entities in the target information and generate an entity graph.

[0136] Specifically, by analyzing the matching degree between various entities (or terms), a graph database corresponding to each entity can be established, resulting in an entity graph (or knowledge graph). Through the entity graph, connections between unstructured data can be quickly established, allowing for the identification of terms in unstructured data corresponding to the target search terms during retrieval, and further identification of related terms based on the entity graph.

[0137] Step S304: Based on the specific sources of local business system data and external data, add corresponding source tags to structured data and unstructured data respectively.

[0138] Specifically, based on the establishment of structured data copies and unstructured data entity maps, further sorting and differentiation can be made based on their sources to further improve the accuracy of retrieval.

[0139] Furthermore, such as Figure 3e The diagram shown is a flowchart of the method for adding data source tags. The specific methods for adding source tags include:

[0140] Step S3041: Add source annotations corresponding to the structured data to the structured data copy.

[0141] Specifically, it is usually relatively easy to determine whether structured data originates from a local business system or an external business system (this can be directly determined by the database shards and tables where it is stored). Therefore, its source can be directly added to the structured data copy in the form of a tag.

[0142] Step S3042: In the sub-data within the same structured data copy, set the corresponding priority label according to the data source.

[0143] Specifically, based on the different sources of the structured data, corresponding priorities can be preset so that data from higher-priority sources is pushed first during retrieval. The specific priority settings are generally based on the priorities pre-configured by the developers.

[0144] After determining the priority, the priority is added to the corresponding data (or sub-data) of the corresponding structured copy.

[0145] Step S3043: In the graph database of unstructured data, add the source label corresponding to each entity.

[0146] Specifically, in graph databases, unstructured data can also be identified by the external business systems that push it when it is acquired, and its corresponding source can be determined. Then, the source can be used as a tag and added to the entries in the entity graph.

[0147] Step S305: Based on the file format of unstructured data, add format tags to the unstructured data, and obtain the business system retrieval thesaurus based on the structured data copy, entity graph, source tags and format tags.

[0148] Specifically, for unstructured data, further format tags can be added based on its specific format, such as text files, emails, mobile data, or media files, so that the corresponding files and data can be accurately retrieved when needed.

[0149] Through steps S302 to S305, different tags are added to the data in the local business system and the external data, and they are highly integrated through structured data copies and entity graphs to obtain the business system retrieval thesaurus, thereby strengthening the correlation between data and effectively ensuring the accuracy and efficiency of obtaining the corresponding data during retrieval.

[0150] Step S306: In response to the obtained search request, determine the corresponding target search term and request object.

[0151] Step S307: Based on the search term library corresponding to the request object and the pre-built business system search term library, perform word segmentation on the target search terms respectively.

[0152] The search term library corresponding to the request object is built based on the search history of the request object, while the search term library of the business system is built based on data from the local business system and data from external databases.

[0153] Specifically, steps S306 to this step are related to... Figure 2 Steps S201 to S202 in the illustrated embodiment are the same and will not be repeated here.

[0154] Step S308: Based on the word segmentation results retrieved from the business system's thesaurus, perform an inverted index on the inventory data.

[0155] Specifically, the target search terms are segmented by searching the thesaurus through the business system. This involves matching the terms or tags corresponding to the target search terms from the tags, structured data copies, and entity graphs obtained from the aforementioned data aggregation platform. The target search terms are then split based on these terms or tags to obtain the segmentation results.

[0156] The entities corresponding to these terms or tags are likely the data needed for retrieval. Since this type of data is usually abundant, it's necessary to build inverted indexes for these entities to quickly locate and retrieve the corresponding data. For example, searching for "steel" in a construction business system might yield thousands of results. Therefore, an inverted index needs to be built to select the data to be pushed (or returned) as the retrieval results based on further configuration.

[0157] Step S309: Based on the priority tags contained in the structured data copy in the inventory data and the word segmentation results corresponding to the business system's search thesaurus, obtain the word segmentation domain priority.

[0158] Specifically, a simple inverted index may contain a large amount of data, far exceeding the data volume required by the requesting object. Therefore, it is necessary to select the parts that need to be pushed first. Here, we can use the priority tags obtained in the previous steps to divide the words or tags matched during word segmentation into different priorities, namely word segmentation domain priorities, in combination with the inverted index to determine the priority of the push.

[0159] Step S310: Based on the inverted index and word segmentation priority, establish a retrieval index.

[0160] Specifically, combining the inverted index with the priority of word segmentation domains yields a unified aggregated index library, i.e., a retrieval index. Through the retrieval index, the address where data matching the target search terms is stored in the retrieval system can be determined, and the priority for pushing data to users can be determined. Data is pushed to the requesting object based on the address of higher-priority data as the retrieval result.

[0161] Step S311: Based on the word segmentation results obtained from the object retrieval thesaurus, determine the retrieval preferences of the requested object.

[0162] Specifically, in addition to searching the index, the search preferences of the requested object can also be considered when determining the search results.

[0163] The terms or tags that exist in the object search term (and those that do not) are obtained by matching the target search term with the object search term database. This is the word segmentation result based on the object search term database. If the target search term contains a term or tag that corresponds to the object search term database, this term or tag is the term or tag preferred when requesting the object search, i.e., the search preference.

[0164] By combining search preferences with search indexes, the search results pushed to the requested object can be further optimized to better meet user needs.

[0165] Step S312: Determine the main retrieval data in the business system corresponding to the object.

[0166] Specifically, in addition to the search preferences of the requested object, the corresponding business system can also be considered, and data in the corresponding business system can be retrieved first to better meet user needs.

[0167] The business system corresponding to the requested object can be a local business system or one of the local business systems (such as multiple local business systems existing simultaneously). If a company is involved in multiple fields, and the data in multiple fields belongs to the local business system, then when personnel in a certain field search, they should prioritize the data in their corresponding field as the primary search data, rather than using the data from all local business systems as the primary search data.

[0168] Step S313: Based on search preferences and entity graphs, determine the data in external business systems associated with the business system corresponding to the object as relevant search data.

[0169] Specifically, after considering the main retrieval data, for data in external business systems, the structured data in the external business systems can be obtained by associating it with the primary key in the structured data copy in the inverted index, and the unstructured data in the external business systems can be obtained through the entity graph. These data can then be used as relevant retrieval data.

[0170] Step S314: Based on the search index, search the main search data and related search data, and display the search results corresponding to the main search data and related search data respectively.

[0171] Specifically, based on the search index and search preferences, searches are performed from the main search data and related search data respectively, and the results obtained are the search results. This ensures coverage of data from both local and external business systems, and allows for targeted differentiation of keywords (i.e., terms) within the business systems (achieved through word segmentation), combined with user behavior (i.e., search preferences), to provide users with highly accurate aggregated search results.

[0172] In some embodiments, Elasticsearch (a distributed full-text search engine based on Lucene) can be used to implement the specific retrieval process.

[0173] Step S315: Obtain user-defined search configuration.

[0174] Specifically, searches can also be performed through user-defined configurations. For example, if the search object is set to search only for a specific business system, then no further expansion or aggregation is needed.

[0175] Step S316: Based on the search configuration and search index, search the local business system data and external data to obtain the search results.

[0176] Specifically, by combining the user-defined search configuration with the search index obtained in the aforementioned steps, targeted search results can also be obtained.

[0177] Steps S315 to S316 are optional steps parallel to steps S311 to S314. Those skilled in the art can choose to execute any step according to their needs.

[0178] The data retrieval method provided in this disclosure collects data from local and external business systems in real time, integrates and processes the data, and then determines the retrieval results based on the retrieval needs of the requesting object and its retrieval history. This enables accurate retrieval of both local and external business data, effectively ensuring the accuracy and reliability of the retrieval results, while also guaranteeing the timeliness of the retrieved external data. Furthermore, by integrating with user behavior, it improves the satisfaction level of the retrieval results.

[0179] Figure 4 This is a schematic diagram of the structure of a data retrieval device provided in one embodiment of the present disclosure. Figure 4 As shown, the data retrieval device 400 includes: an acquisition module 410, a word segmentation module 420, an indexing module 430, and an output module 440. Wherein:

[0180] The acquisition module 410 is used to determine the corresponding target search term and request object in response to the acquired search request;

[0181] The word segmentation module 420 is used to segment the target search terms based on the search term library corresponding to the request object and the pre-built business system search term library. The search term library corresponding to the request object is built based on the search history of the request object, and the business system search term library is built based on the data in the local business system and the data in the external database.

[0182] Index module 430 is used to build a corresponding search index based on the word segmentation processing results and pre-set priority tags of the business system's retrieval thesaurus.

[0183] The output module 440 is used to establish a corresponding search index based on the word segmentation results and pre-set priority tags of the business system's search terminology.

[0184] Optionally, the word segmentation module 420 includes obtaining the business system retrieval lexicon through the following methods: acquiring local business system data and external data collected from external databases; obtaining corresponding structured data copies based on the structured data in the local business system data and external data; obtaining corresponding entity graphs based on the unstructured data in the local business system data and external data, the entity graphs being used to represent the relationships between the local business system data and external data; adding corresponding source tags to the structured data and unstructured data respectively based on the specific sources of the local business system data and external data; and adding format tags to the unstructured data based on the file format of the unstructured data.

[0185] Optionally, the word segmentation module 420 is specifically used to: establish a primary key association between the structured data and the local business system data; integrate the external data based on the primary key association; and combine the integration result with the local business system data based on the primary key association to generate a copy of the structured data.

[0186] Optionally, the word segmentation module 420 is specifically used to: extract target information from unstructured data, including entities, entity attributes, and classification labels; based on the target information, obtain the matching degree between unstructured data through correlation analysis; extract the corresponding words of entities in the target information and mark the matching degree on the words; based on the matching degree, establish a graph database corresponding to the entities in the target information and generate an entity graph.

[0187] Optionally, the word segmentation module 420 is specifically used to: add source annotations corresponding to the structured data in the structured data copy; set corresponding priority tags according to the data source in the sub-data in the same structured data copy; and add source tags corresponding to each entity in the graph database of unstructured data.

[0188] Optionally, the word segmentation module 420 includes collecting external data in the following ways: if the external data is structured data, monitor the data changes in the external database of the corresponding external business system in real time; collect the structured data in the external data according to the type of data content corresponding to the data changes; if the external data is unstructured data, obtain the push request from the external business system; and obtain the external data corresponding to the external business system based on the push request.

[0189] Optionally, the word segmentation module 420 is specifically used to: if the data content is a single table data change, send a pull task to the external business system to extract the single table data information; if the data content is continuously incremental change information, send a subscription task to the external business system to continuously obtain real-time incremental change information through subscription; if the data content is existing data that has not been collected, pull all external data information.

[0190] Optionally, the word segmentation module 420 is specifically used to collect unstructured data from external data to a local storage location by setting a transmission method; and to extract structured data information from the unstructured data in the local storage location by setting a collection method.

[0191] Optionally, the word segmentation module 420 is also used to: after obtaining external data based on a push request, perform legality verification on the obtained external data; perform desensitization processing on the business-sensitive data in the external data that has passed the legality verification; and save the external data to the corresponding database shard or table based on the external business system corresponding to the desensitized external data.

[0192] Optionally, the index module 430 is specifically used to: perform an inverted index on the inventory data based on the word segmentation processing results corresponding to the business system's retrieval thesaurus; obtain the word segmentation domain priority based on the priority tags contained in the structured data copies in the inventory data and the word segmentation processing results corresponding to the business system's retrieval thesaurus; and establish a retrieval index based on the inverted index and the word segmentation domain priority.

[0193] Optionally, the output module 440 is specifically used to: determine the search preferences of the requested object based on the word segmentation results obtained from the object search thesaurus; determine the data in the business system corresponding to the object as the main search data; determine the data in the external business system associated with the business system corresponding to the object as the relevant search data based on the search preferences and entity graph; and search the main search data and relevant search data based on the search index, and display the search results for the main search data and relevant search data respectively.

[0194] Optionally, the output module 440 is specifically used to obtain user-defined search configurations; and based on the search configurations and search indexes, to search local business system data and external data to obtain search results.

[0195] In this embodiment, the data retrieval device, through the combination of various modules, can solve the problem in the prior art that search results cannot meet user expectations when searching for diversified data. The obtained retrieval results can simultaneously integrate proprietary data in the local business system and data in external databases, improve the retrieval efficiency and accuracy of data in external databases, and thus improve the accuracy of retrieval results.

[0196] Figure 5 This is a schematic diagram of the structure of a control device provided in one embodiment of the present disclosure, as shown below. Figure 5 As shown, the control device 500 includes a memory 510 and a processor 520.

[0197] The memory 510 stores a computer program that can be executed by at least one processor 520. This computer program is executed by at least one processor 520 to enable the control device to implement the data retrieval method provided in any of the above embodiments.

[0198] The memory 510 and the processor 520 can be connected via a bus 530.

[0199] The relevant explanations can be understood by referring to the corresponding descriptions and effects in the method embodiments, and will not be repeated here.

[0200] One embodiment of this disclosure provides a computer-readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the data retrieval method provided in any of the above method embodiments.

[0201] The computer-readable storage medium can be ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0202] One embodiment of this disclosure provides a computer program product comprising computer-executable instructions that, when executed by a processor, are used to implement the data retrieval method provided in any of the above embodiments.

[0203] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0204] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the disclosure herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope of this application is indicated by the claims.

[0205] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope.

Claims

1. A data retrieval method, characterized in that, The data retrieval method includes: In response to the received search request, determine the corresponding target search terms and the request object; Acquire local business system data and external data collected from external databases, wherein the external data includes structured data and unstructured data; If the external data is structured data, a corresponding structured data copy is obtained based on the structured data in the local business system and the external data. The structured data copy is used to store the correspondence between the structured data in the local business system and the structured data in the external data and the primary key of the database in the data acquisition platform. If the external data is unstructured data, extract the target information of the unstructured data, which includes entities, entity attributes, and classification labels; Based on the target information, the matching degree between the unstructured data is obtained through correlation analysis; Extract the corresponding terms from the target information and mark the matching degree on the terms; Based on the matching degree, a graph database corresponding to the entities in the target information is established, and an entity graph is generated. The entity graph is used to represent the relationship between local business system data and external data. Based on the specific sources of the local business system data and the external data, corresponding source tags are added to the structured data and unstructured data respectively; Based on the file format of unstructured data, format tags are added to the unstructured data, and a business system retrieval thesaurus is constructed based on the structured data copy, entity graph, source tags, and format tags; Based on the object search term library corresponding to the requested object and the pre-built business system search term library, the target search terms are segmented respectively. The object search term library corresponding to the requested object is built based on the search history of the requested object. Based on the word segmentation results and pre-set priority tags of the retrieval lexicon in the business system, a corresponding retrieval index is established; Based on the business system corresponding to the requested object, the word segmentation results obtained from the object search thesaurus, and the search index, the search results are obtained.

2. The data retrieval method according to claim 1, characterized in that, The process of obtaining a corresponding structured data copy based on the structured data from the local business system data and the external data includes: Establish a primary key association between the structured data and the database corresponding to the local business system data; Based on the primary key association, the exogenous data is integrated; The integrated result is combined with the local business system data based on primary key association to generate the structured data copy.

3. The data retrieval method according to claim 1, characterized in that, Based on the specific sources of the local business system data and external data, corresponding source tags are added to the structured and unstructured data respectively, including: Add the source label corresponding to the structured data to the copy of the structured data; Within the sub-data in the same structured data copy, set corresponding priority labels based on the data source; In the graph database of the unstructured data, add a source tag corresponding to each entity.

4. The data retrieval method according to any one of claims 1 to 3, characterized in that, The exogenous data was collected in the following manner: If the external data is structured data, monitor the data changes in the external database of the corresponding external business system in real time; Based on the type of data content corresponding to the data changes, collect structured data from external data sources; If the external data is unstructured data, obtain the push request from the external business system; Based on the push request, obtain the external data corresponding to the external business system.

5. The data retrieval method according to claim 4, characterized in that, The step of collecting structured data from external data based on the type of data content corresponding to data changes includes: If the data content is a single table data change, a pull task is sent to the external business system to extract the single table data information; If the data content is continuously incrementally changing information, a subscription task is sent to the external business system to continuously obtain real-time incremental change information through subscription. If the data content is existing data that has not been collected before, retrieve all of the external data information.

6. The data retrieval method according to claim 4, characterized in that, The step of obtaining the exogenous data based on the push request includes: Unstructured data from external data sources is collected and stored locally based on a set transmission method. Structured data information is extracted from unstructured data in the local storage location by setting the acquisition method.

7. The data retrieval method according to claim 4, characterized in that, After obtaining the exogenous data based on the push request, the process further includes: Perform legality verification on the acquired external data; De-identify sensitive business data in externally sourced data that has passed legality verification; Based on the external business system corresponding to the de-identified external data, the external data is saved to the corresponding database shard or table.

8. The data retrieval method according to any one of claims 1 to 3, characterized in that, The process of establishing a corresponding search index based on the word segmentation results and pre-set priority tags corresponding to the business system's search terminology database includes: Based on the word segmentation results of the lexicon retrieved by the business system, an inverted index is created for the inventory data. Based on the priority tags contained in the structured data copy of the inventory data and the word segmentation results corresponding to the lexicon retrieved by the business system, the word segmentation domain priority is obtained; The retrieval index is established based on the inverted index and the word segmentation priority.

9. The data retrieval method according to any one of claims 1 to 3, characterized in that, The search results obtained based on the business system corresponding to the request object, the word segmentation results obtained from the object search thesaurus, and the search index include: Based on the word segmentation results obtained from the object retrieval thesaurus, the retrieval preferences of the requested object are determined; The data in the business system corresponding to the object is determined to be the main retrieval data; Based on the search preferences and the entity graph, data from external business systems associated with the business system corresponding to the object are identified as relevant search data. Based on the search index, the main search database and the related search database are searched, and the data obtained from the search in the main search database and the related search database are displayed respectively to obtain the search results.

10. The data retrieval method according to any one of claims 1 to 3, characterized in that, The search results obtained based on the business system corresponding to the request object, the word segmentation results obtained from the object search thesaurus, and the search index include: Retrieve user-defined search configurations; Based on the search configuration and the search index, local business system data and external data are searched to obtain the search results.

11. A data retrieval device, characterized in that, The data retrieval device includes: The acquisition module is used to determine the corresponding target search terms and request objects in response to the acquired search request; The word segmentation module is used to obtain data from local business systems and external data collected from external databases. The external data includes structured data and unstructured data. If the external data is structured data, a corresponding structured data copy is obtained based on the structured data in the local business system and the external data. The structured data copy is used to store the correspondence between the structured data in the local business system and the structured data in the external data and the primary key of the database in the data acquisition platform. If the external data is unstructured data, extract the target information of the unstructured data, which includes entities, entity attributes, and classification labels; Based on the target information, the matching degree between the unstructured data is obtained through correlation analysis; Extract the corresponding terms from the target information and mark the matching degree on the terms; Based on the matching degree, a graph database corresponding to the entities in the target information is established, and an entity graph is generated. The entity graph is used to represent the relationship between local business system data and external data. Based on the specific sources of the local business system data and the external data, corresponding source tags are added to the structured data and unstructured data respectively; Based on the file format of unstructured data, format tags are added to the unstructured data, and a business system retrieval thesaurus is constructed based on the structured data copy, entity graph, source tags, and format tags; Based on the object search term library corresponding to the requested object and the pre-built business system search term library, the target search terms are segmented respectively. The object search term library corresponding to the requested object is built based on the search history of the requested object. The index module is used to build a corresponding search index based on the word segmentation results and pre-set priority tags of the business system's retrieval thesaurus. The output module is used to build a corresponding search index based on the word segmentation results of the business system's search thesaurus and the pre-set priority tags.

12. A control device, characterized in that, include: At least one processor; and a memory communicatively connected to the at least one processor; The memory stores instructions executable by the at least one processor, which, when executed by the at least one processor, cause the control device to perform the data retrieval method as described in any one of claims 1 to 10.

13. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the data retrieval method as described in any one of claims 1 to 10.

14. A computer program product, characterized in that, The computer program product includes computer execution instructions, which, when executed by a processor, are used to implement the data retrieval method as described in any one of claims 1 to 10.

Images