Method for constructing education index database, query method, device and electronic equipment

Abstract

This invention relates to the field of artificial intelligence technology, and discloses a method, query method, apparatus, and electronic device for constructing an educational index database. The method includes: acquiring unstructured data corresponding to an educational management system, as well as a first identifier and first metadata corresponding to the unstructured data; performing vectorization processing on the unstructured data to obtain feature vectors corresponding to the unstructured data; storing the feature vectors in a pre-constructed initial vector library to obtain a target vector library; and storing the first metadata in a pre-constructed initial relational database to obtain a target relational database; and associating the feature vectors in the target vector library and the first metadata in the target relational database based on the first identifier to obtain the educational index database. By implementing the method of this application, the problem of lacking an indexing mechanism that supports cross-modal data querying can be solved.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, specifically to a method for constructing an educational index database, a query method, a device, and an electronic device. Background Technology

[0002] With the continuous advancement of educational informatization, various educational management systems, such as those for academic affairs, scientific research, personnel, attendance, and student evaluation, have become widely used. During their operation, these educational management systems have not only accumulated a large amount of structured data but also a wealth of unstructured data, such as real-life images of teaching venues, images of scientific research equipment, trend charts in statistical analysis reports, and scanned copies of various award certificates.

[0003] Currently, artificial intelligence technologies, such as large language models or AI agents, are typically used to analyze the accumulated data to generate analytical reports addressing relevant educational management issues. However, there is a lack of indexing mechanisms that support cross-modal data queries. Specifically, when users input natural language queries such as "find a chart showing the growth trend of research funding over the past three years," it is difficult to directly retrieve the corresponding chart. This deficiency results in the generated analytical reports lacking accurate visual evidence, thus affecting the accuracy and intuitiveness of educational management analysis. Summary of the Invention

[0004] This invention provides a method for constructing an educational index database, a query method, a device, and an electronic device to solve the problem of the lack of an indexing mechanism that supports cross-modal data querying.

[0005] In a first aspect, the present invention provides a method for constructing an educational index database. The method includes: acquiring unstructured data corresponding to an educational management system, as well as a first identifier and first metadata corresponding to the unstructured data; performing vectorization processing on the unstructured data to obtain feature vectors corresponding to the unstructured data; storing the feature vectors in a pre-constructed initial vector library to obtain a target vector library; and storing the first metadata in a pre-constructed initial relational database to obtain a target relational database; and associating the feature vectors in the target vector library and the first metadata in the target relational database based on the first identifier to obtain the educational index database.

[0006] The method for constructing an educational index database provided in this embodiment involves vectorizing unstructured data to obtain feature vectors, which are then stored in an initial vector library to form a target vector library. During subsequent retrieval, the target vector library can be used to match user query information with unstructured data, identifying the unstructured data that matches the user's query. Furthermore, this application uses the first metadata of the unstructured data to form a target relational database and associates it with the target vector library. After identifying the unstructured data that matches the user's query, the first metadata corresponding to the unstructured data can be obtained from the target relational database. Based on this first metadata, unstructured data can be retrieved from the educational management system, enabling cross-modal data querying and solving the problem of a lack of indexing mechanisms supporting cross-modal data querying. Moreover, the educational index database of this application does not store unstructured data, saving storage costs and avoiding the risks associated with data transfer, achieving zero-intrusion management of the original data.

[0007] In one optional implementation, the process of obtaining the first metadata corresponding to the unstructured data includes: pulling the storage path and file name of the unstructured data in the education management system; and obtaining the first metadata corresponding to the unstructured data by fusing the first identifier, file name and storage path corresponding to the unstructured data.

[0008] By integrating the first identifier, storage path, and file name corresponding to unstructured data into structured first metadata, unstructured data can be organized, queried, and maintained in a unified manner, thus improving the efficiency of retrieving unstructured data.

[0009] In an optional implementation, the method further includes: obtaining the data table corresponding to the education management system and the second metadata and second identifier corresponding to the data table; storing the second metadata in the target relational database to obtain the updated target relational database; associating the data table with other data tables and education management statistical indicators in the pre-constructed initial graph database to obtain the target graph database; and associating the second metadata in the updated target relational database with the data table in the target graph database based on the second identifier to obtain the updated education index database.

[0010] Based on the second identifier corresponding to the data table, the updated second metadata in the target relational database is associated with the data table in the target graph database, realizing the integration of the association between the structured storage second metadata and the graph storage. In this way, the updated education index database can improve data retrieval, tracing and analysis capabilities, and also reduce the complexity of data management.

[0011] In one optional implementation, the process of obtaining the second metadata corresponding to the data table includes: retrieving the database name, data table name, and field names of the data table within the education management system; and obtaining the second metadata based on the fusion of the second identifier, database name, data table name, and field names.

[0012] By integrating the secondary identifier, database name, data table name, and field names corresponding to the data table into structured secondary metadata, it is possible to achieve structured, standardized, and unified management of relevant information of the data table in the education management system, providing a reliable metadata foundation for subsequent data retrieval.

[0013] In one optional implementation, the data table is associated with other data tables and educational management statistical indicators in a pre-constructed initial graph database to obtain a target graph database. This includes: updating the nodes corresponding to the data table to the initial graph database to obtain a candidate graph database; based on the foreign key information corresponding to the data table, constructing a first association relationship between the nodes corresponding to the data table and the nodes corresponding to other data tables in the candidate graph database; and based on the query statement corresponding to the educational management statistical indicators, constructing a second association relationship between the nodes corresponding to the educational management statistical indicators and the nodes corresponding to the data tables in the candidate graph database to obtain the target graph database.

[0014] Based on foreign key information, the first association between data tables and other data tables can be accurately established in the candidate graph database. Furthermore, based on the query statements corresponding to education management statistical indicators, the second association between data tables and education management statistical indicators can be accurately established in the candidate graph database. This allows information originally scattered across different databases and data tables to be uniformly expressed in graph form, thus presenting the inherent connections between data more intuitively. Subsequently, users can start from the nodes corresponding to education management statistical indicators, connect them to the corresponding data tables through the graph database, and query the second metadata corresponding to the data tables from the updated target relational database based on the second identifier of the data tables. Then, based on the second metadata, users can locate the data in the data tables within the education management system and calculate the specific values ​​corresponding to the education statistical indicators.

[0015] In an optional implementation, the method further includes: adding the nodes corresponding to each field in the second metadata to the target graph database, and establishing the association between the nodes corresponding to each field in the target graph database and the nodes corresponding to the data tables to obtain the updated target graph database; determining the semantic similarity between the first target node and the second target node in the updated target graph database, wherein the first target node is the node corresponding to any field in the second metadata, and the second target node is the node corresponding to any field in other data tables; if the semantic similarity is greater than or equal to a preset similarity threshold, then the first target node and the second target node are merged into the same target node, and the node information of the target node is configured as the field information corresponding to the first target node and the field information corresponding to the second target node.

[0016] By calculating the semantic similarity between fields added to the target graph database and fields already in the target graph database, highly similar nodes are automatically merged. This unifies heterogeneous fields into a single target node, resolving the issue of heterogeneous field naming in the education management system. It also enhances the field-level association capabilities of the graph database, significantly improving data consistency, retrieval efficiency, and analytical depth. Furthermore, regardless of the user's preferred expression method, the system can correctly retrieve the data corresponding to that field.

[0017] Secondly, the present invention provides a query method for an educational index database, wherein the educational index database is constructed according to the construction method of the educational index database according to the first aspect or any corresponding embodiment thereof. The method includes: receiving query information; parsing the query information to determine the query type corresponding to the query information; if the query type represents querying unstructured data, then performing vector matching between the query vector corresponding to the query information and the feature vector in the target vector library to obtain a target feature vector matching the query information; querying a target relational database based on a first identifier corresponding to the target feature vector to obtain first metadata corresponding to the target feature vector; and retrieving target unstructured data corresponding to the target feature vector from an educational management system based on the first metadata corresponding to the target feature vector to obtain the target unstructured data corresponding to the query information.

[0018] The query method for the educational index database provided in this embodiment obtains the target feature vector by matching the query vector corresponding to the query information with the feature vector in the target vector library. Then, it determines the first metadata of the target feature vector with the help of the target relational database. With the help of the first metadata, the target unstructured data corresponding to the query information can be accurately retrieved, thereby realizing cross-modal data query.

[0019] In an optional implementation, the method further includes: if the query type represents a query for educational management statistical indicators, then semantically matching the query statistical indicators in the query information with the educational management statistical indicators in the target graph database to determine the target educational management statistical indicators corresponding to the query information; querying the target data table associated with the target educational management statistical indicators from the target graph database; querying a new target relational database based on the second identifier corresponding to the target data table to obtain the second metadata corresponding to the target data table; pulling the target data table from the educational management system based on the second metadata, and processing the data in the target data table based on the query statement corresponding to the target educational management statistical indicators to obtain the numerical values ​​corresponding to the query information.

[0020] By using the target graph database, the data table corresponding to the query statistical indicators in the query information can be accurately identified. Furthermore, by using the updated target relational database, the secondary metadata such as the education management system and database name corresponding to the data table can be accurately identified. Then, by using the secondary metadata, the data corresponding to the data table in the education management system can be located. Finally, by using the query statement corresponding to the target education management statistical indicators, the value corresponding to the query information can be accurately determined.

[0021] Thirdly, the present invention provides an apparatus for constructing an educational index database, the apparatus comprising: a first acquisition module for acquiring unstructured data corresponding to an educational management system, and a first identifier and first metadata corresponding to the unstructured data; a first processing module for vectorizing the unstructured data to obtain feature vectors corresponding to the unstructured data; a first storage module for storing the feature vectors in a pre-constructed initial vector library to obtain a target vector library, and storing the first metadata in a pre-constructed initial relational database to obtain a target relational database; and a first association module for associating the feature vectors in the target vector library and the first metadata in the target relational database based on the first identifier to obtain the educational index database.

[0022] Fourthly, the present invention provides an electronic device, comprising: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to perform the method for constructing an educational index database according to the first aspect or any corresponding embodiment described above.

[0023] Fifthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to execute the method for constructing an educational index database according to the first aspect or any corresponding embodiment thereof.

[0024] In a sixth aspect, the present invention provides a computer program product, including computer instructions for causing a computer to execute the method for constructing an educational index database according to the first aspect or any corresponding embodiment described above. Attached Figure Description

[0025] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0026] Figure 1 This is a schematic diagram of an application scenario according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the first process of constructing an educational index database according to an embodiment of the present invention; Figure 3 This is a schematic diagram of a second process for constructing an educational index database according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the association relationships in the target graph database according to an embodiment of the present invention; Figure 5 This is a schematic diagram of the first type of query method for an educational index database according to an embodiment of the present invention; Figure 6 This is a flowchart illustrating a specific method for constructing and querying an educational index database according to an embodiment of the present invention. Figure 7 This is a structural block diagram of an apparatus for constructing an educational index database according to an embodiment of the present invention; Figure 8 This is a structural block diagram of a query device for an educational index database according to an embodiment of the present invention; Figure 9 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] It is understood that before using the technical solutions disclosed in the various embodiments of the present invention, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in the present invention and their authorization should be obtained in accordance with relevant laws and regulations through appropriate means.

[0029] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0030] With the continuous advancement of educational informatization, various educational management systems, such as those for academic affairs, scientific research, personnel, attendance, and student evaluation, have become widely used. In the process of operation, these educational management systems have not only accumulated a large amount of structured and semi-structured data, but also accumulated a wealth of unstructured data, such as real-life images of teaching venues, images of scientific research equipment, trend charts in statistical analysis reports, and scanned copies of various award certificates.

[0031] Currently, artificial intelligence technologies, such as large language models or AI agents, are typically used to analyze the accumulated data to generate analytical reports addressing relevant educational management issues. However, the following problems still exist in the data analysis process: 1. The definition and calculation methods of the same educational management statistical indicator are inconsistent across different educational management systems. For example, for the educational management statistical indicator of "number of full-time teachers," the academic affairs department may only include personnel undertaking teaching duties, while the human resources department may include both teaching duties and full-time research personnel. Similarly, the statistical methods for the educational management statistical indicator of "number of research achievements" differ between the research office and the finance department. Furthermore, the same concept often appears in multiple ways within the education sector; for example, "student academic warning" may be called "academic risk" or "academic alert." This lack of a unified semantic system makes it easy for intelligent retrieval and analysis results to contain omissions or ambiguities.

[0032] 2. Currently, there is a lack of indexing mechanisms for visual materials based on natural language queries. Specifically, when users input natural language queries such as "find charts showing the growth trend of research funding over the past three years," it is difficult to directly retrieve corresponding visual materials. This deficiency results in the generated analysis reports lacking accurate visual evidence, thus affecting the accuracy and intuitiveness of educational management analysis.

[0033] To address the aforementioned problems, this application first generates metadata corresponding to structured, semi-structured, and unstructured data for the education management system. Then, based on this metadata, a relational database is constructed, forming a vector library for unstructured data and a graph database for structured and semi-structured data. The relational database and vector library are linked using identifiers corresponding to unstructured data, and the relational database and graph database are linked using identifiers corresponding to structured or semi-structured data. Subsequently, during queries, the vector library enables semantic matching between unstructured data and user query information, and the relational database allows for rapid retrieval of unstructured data corresponding to the query information from the relevant education management system, achieving cross-modal data querying. Similarly, the graph database and relational database can quickly match the standardized education management statistical indicators (i.e., the target education management statistical indicators) corresponding to the education management statistical indicators in the query information, as well as the corresponding structured or semi-structured data, avoiding situations where inaccurate descriptions of the education management statistical indicators in the query information lead to inaccurate values ​​of the education management statistical indicators in the final retrieved query information.

[0034] As an optional application scenario of this invention, such as Figure 1 As shown, the system for building this educational index database may include at least one terminal device and at least one server. Figure 1 The system is illustrated in the example, which includes a computer 101, a mobile terminal 102, and a server 103, and the terminal devices such as the computer 101 and the mobile terminal 102 are connected to the server 103 through a network 110.

[0035] Specifically, the terminal device can be a smartphone, tablet, laptop, PDA, desktop computer, game console, smart TV, smart wearable device, in-vehicle terminal, VR (Virtual Reality) device, AR (Augmented Reality) device, etc. Server 103 can be a standalone physical server, a server cluster, a distributed system, or a cloud server providing cloud services. Network 110 can be a wired or wireless network, examples of which include, but are not limited to, the Internet, corporate intranet, local area network, wide area network, mobile communication network, and combinations thereof.

[0036] According to an embodiment of the present invention, an embodiment of a method for constructing an educational index database is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0037] This embodiment provides a method for constructing an educational index database, which can be used in terminal devices. Figure 2 This is a flowchart of a method for constructing an educational index database according to an embodiment of the present invention, such as... Figure 2 As shown, the process includes the following steps: Step S201: Obtain the unstructured data corresponding to the education management system, as well as the first identifier and first metadata corresponding to the unstructured data.

[0038] Educational management systems can be the management systems mentioned above, such as those for academic affairs, research, personnel, attendance, and student evaluation. Examples include academic affairs systems, research management systems, personnel systems, attendance systems, and student evaluation systems.

[0039] Unstructured data refers to data that cannot be stored and organized according to a fixed format. It typically includes various types such as text, images, audio, and video. Its content structure is complex and difficult to directly query and analyze using traditional relational databases. The unstructured data in this application may include, but is not limited to, real-life scenes of teaching venues, images of scientific research equipment, trend charts in statistical analysis reports, scanned copies of various award certificates, and scanned copies of questionnaires, etc.

[0040] The first identifier is used to uniquely identify an unstructured data. For example, the first identifier can be the identifier (ID) information corresponding to the unstructured data. The ID information can be encoded according to a preset encoding method. Of course, the ID information can also be the MD5 (Message-Digest Algorithm 5) value obtained by hashing the description information corresponding to the unstructured data.

[0041] Metadata can be data about data, describing the data's attributes, characteristics, storage location, and management information. In this context, the primary metadata for unstructured data can describe the educational management system containing the unstructured data, as well as the file name and storage path of the unstructured data within that system.

[0042] As a specific example, the educational index database construction system of this application can establish a communication connection with an educational management system to obtain the corresponding unstructured data and the corresponding first metadata from the educational management system. Alternatively, a communication connection can be established between the educational index database construction system and the database corresponding to the educational management system to obtain the unstructured data and the corresponding first metadata from the educational management system.

[0043] Step S202: Vectorize the unstructured data to obtain the feature vectors corresponding to the unstructured data.

[0044] If the unstructured data is text data such as reports or documents, it can be processed based on a pre-trained language model, a combination of Term Frequency-Inverse Document Frequency (TF-IDF) and vector mapping, or a combination of Bag of Words model and vector transformation to obtain the feature vector corresponding to the unstructured data.

[0045] If the unstructured data here is image data such as photos of equipment, pictures of teaching scenes, charts, etc., it can be processed based on a pre-trained image embedding model, Transformer model, or a combination of feature extraction and vector transformation to obtain the feature vector corresponding to the unstructured data.

[0046] If the unstructured data here is audio data such as classroom recordings, meeting recordings, or voice evaluations, then a combination of voice feature extraction and deep learning models can be used to process the unstructured data and obtain the feature vector corresponding to the unstructured data.

[0047] Step S203: Store the feature vectors into a pre-built initial vector library to obtain the target vector library, and store the first metadata into a pre-built initial relational database to obtain the target relational database.

[0048] Relational databases can store structured data in the form of tables, establishing relationships between tables through rows, columns, and primary / foreign keys. Vector databases are databases used to store and query vectors, which are derived from the vectorization of text, speech, images, video, etc.

[0049] As a specific example, the feature vectors corresponding to unstructured data and the first identifier can be stored in an initial vector library to obtain a target vector library, and the feature vectors corresponding to unstructured data and the first metadata can be stored in an initial relational database to obtain a target relational database.

[0050] Step S204: Based on the first identifier, associate the feature vectors in the target vector library with the first metadata in the target relational database to obtain the education index database.

[0051] As a specific example, a relational database can be constructed. This relational database can be used to store the correspondence between the first identifier, the identifier information of the feature vector corresponding to the unstructured data in the target vector library, and the identifier information of the first metadata corresponding to the unstructured data in the target relational database. In this way, the first identifier is used to associate the feature vector in the target vector library with the first metadata in the target relational database, resulting in an educational index database that includes the relational database, the target vector library, and the target relational database.

[0052] After the educational index database of this application is completed, it can provide a unified service interface to the outside world. With the help of the unified service interface, the indexing capabilities of the educational index database can be opened to upper-layer applications such as question-and-answer systems and automatic report generation systems, supporting keyword retrieval, semantic retrieval and graph reasoning, etc., to ensure the accuracy, authority and sustainable evolution capability of educational management data applications.

[0053] The method for constructing an educational index database provided in this embodiment involves vectorizing unstructured data to obtain feature vectors, which are then stored in an initial vector library to form a target vector library. During subsequent retrieval, the target vector library can be used to match user query information with unstructured data, identifying the unstructured data that matches the user's query. Furthermore, this application uses the first metadata of the unstructured data to form a target relational database and associates it with the target vector library. After identifying the unstructured data that matches the user's query, the first metadata corresponding to the unstructured data can be obtained from the target relational database. Based on this first metadata, unstructured data can be retrieved from the educational management system, enabling cross-modal data querying and solving the problem of a lack of indexing mechanisms supporting cross-modal data querying. Moreover, the educational index database of this application does not store unstructured data, saving storage costs and avoiding the risks associated with data transfer, achieving zero-intrusion management of the original data.

[0054] This embodiment provides a method for constructing an educational index database, which can be used in terminal devices. Figure 3 This is a flowchart of a method for constructing an educational index database according to an embodiment of the present invention, such as... Figure 3 As shown, the process includes the following steps: Step S301: Obtain the unstructured data corresponding to the education management system, as well as the first identifier and first metadata corresponding to the unstructured data. For details, please refer to [link to relevant documentation]. Figure 2 Step S201 of the illustrated embodiment will not be described again here.

[0055] In some optional implementations, the process of obtaining the first metadata corresponding to the unstructured data includes: Step a1: Retrieve the storage path and filename of the unstructured data in the education management system.

[0056] Step a2: Based on the fusion of the first identifier, file name and storage path corresponding to the unstructured data, the first metadata corresponding to the unstructured data is obtained.

[0057] As shown above, the educational index database construction system of this application can establish a communication connection with the educational management system. Subsequently, it can access the metadata of unstructured data in the educational management system, thereby pulling the storage path of unstructured data in the educational management system, such as Uniform Resource Identifier (URI), file name, creation time, etc.

[0058] As a concrete example, a mapping relationship can be established between the first identifier, the file name, and the storage path to associate and store these three elements. Thus, using the first identifier as the index key, the scattered file names and storage paths can be integrated into a structured record, thereby generating the first metadata corresponding to the unstructured data.

[0059] Furthermore, after constructing the primary metadata for unstructured data, a real-time monitoring mechanism can be established to continuously capture various changes to the unstructured data within the education management system throughout its lifecycle, including but not limited to additions, modifications, deletions, path adjustments, or attribute updates. Upon detecting these changes, a process for automatically updating the primary metadata can be triggered, ensuring that the primary metadata remains consistent with the actual state of the unstructured data within the education management system.

[0060] By integrating the first identifier, storage path, and file name corresponding to unstructured data into structured first metadata, unstructured data can be organized, queried, and maintained in a unified manner, thus improving the efficiency of retrieving unstructured data.

[0061] Step S302 involves vectorizing the unstructured data to obtain the corresponding feature vectors. For details, please refer to [link to relevant documentation]. Figure 2 Step S202 of the illustrated embodiment will not be described again here.

[0062] Step S303 involves storing the feature vectors into a pre-built initial vector library to obtain the target vector library, and storing the first metadata into a pre-built initial relational database to obtain the target relational database. For details, please refer to [link to relevant documentation]. Figure 2 Step S203 of the illustrated embodiment will not be described again here.

[0063] Step S304: Based on the first identifier, associate the feature vectors in the target vector library with the first metadata in the target relational database to obtain the education index database. For details, please refer to [link to relevant documentation]. Figure 2 Step S204 of the illustrated embodiment will not be described again here.

[0064] Step S305: Obtain the data table corresponding to the education management system, as well as the second metadata and second identifier corresponding to the data table.

[0065] Data tables can be structured or semi-structured. Structured data tables are used to store highly organized, fixed-format data that can be directly stored in relational databases; semi-structured data tables are used to store data that falls between structured and unstructured data. Although this type of data does not have a strict two-dimensional table structure, it describes the hierarchy and attributes of the data through specific tags or formats.

[0066] The second identifier here can be used to uniquely identify a data table. This second identifier can be the ID information corresponding to the data table, and the generation method for the ID information can refer to the generation method for ID information corresponding to unstructured data described earlier. The second metadata corresponding to the data table can be used to describe the education management system to which the data table belongs, as well as the database name and table name within that system.

[0067] The method for obtaining the second metadata corresponding to the data table here can refer to the method for obtaining the first metadata corresponding to unstructured data shown above.

[0068] In some optional implementations, the process of obtaining the second metadata corresponding to the data table includes: Step b1: Retrieve the database name, table name, and field names of the data table within the education management system.

[0069] Step b2: Based on the fusion of the second identifier, database name, data table name, and field names, the second metadata is obtained.

[0070] As shown above, the educational index database construction system of this application can establish a communication connection with the educational management system. Subsequently, it can access the metadata of the data table in the educational management system, thereby retrieving the database name, data table name, and field names to which the data table belongs in the educational management system.

[0071] As a concrete example, a mapping relationship can be established between the second identifier and the database name, table name, and field names to link and store these four elements together. Thus, using the second identifier as the index key, the scattered database name, table name, and field names can be integrated into a structured record, thereby generating the second metadata corresponding to the unstructured data.

[0072] In addition, after constructing the secondary metadata for the data tables, a real-time monitoring mechanism can be established for non-data tables within the education management system to continuously capture various changes throughout their lifecycle. These changes include, but are not limited to, additions, modifications, deletions, path adjustments, or updates to table structure, fields, and data types. Upon detecting these changes, a process for automatically synchronizing and updating the secondary metadata can be triggered, ensuring that the secondary metadata always remains consistent with the actual state of the data tables within the education management system.

[0073] By integrating the secondary identifier, database name, data table name, and field names corresponding to the data table into structured secondary metadata, it is possible to achieve structured, standardized, and unified management of relevant information of the data table in the education management system, providing a reliable metadata foundation for subsequent data retrieval.

[0074] Step S306: Store the second metadata in the target relational database to obtain the updated target relational database.

[0075] As a specific example, the second identifier and the second metadata can be stored simultaneously in the target relational database to obtain an updated target relational database. Furthermore, in the actual storage process, the first metadata and the second metadata can reside in the same table in the target relational database. Of course, the first metadata and the second metadata can also reside in different tables in the target relational database; this application does not impose specific limitations on this, and adjustments can be made flexibly according to the actual situation.

[0076] As a concrete example, the table containing the second piece of metadata in the new target relational database can be called a logical unit table. This logical unit table may also include the following fields: Unit_ID (UUID) field, used to represent the primary key; Indicator_ID (Foreign Key) field, used to represent the ID information of the educational statistical indicator associated with the data table; Logic_Type (Enum) field, used to represent the type of query statement corresponding to the associated educational statistical indicator, such as SQL, StoredProcedure, or PythonScript; Script_Content (Text / CLOB) field, used to represent the validated and complete executable code, such as SELECT count(*) FROM t_student WHERE status=1; Param_Schema (JSON), used to represent dynamic parameters in the query statement, such as year and department; Hash_Check (String) field, used to represent the hash value corresponding to the query statement, used for integrity verification; and Version_Tag (String) field, used to represent the version number of the logical unit table.

[0077] As a concrete example, the table containing the first piece of metadata in the target relational database can be called a multimodal anchor table. This multimodal anchor table may also include the following fields: Anchor_ID (UUID) field, used to represent the primary key; Resource_Type (Enum) field, used to represent the resource type of the unstructured data, such as images, audio / video, charts, and text; Source_URI (String) field, used to represent the storage path of the unstructured data on the original file server or object storage, such as oss: / / edu-data / 2024 / doc_01.png; Vector_Ref_ID (String) field, used to store the ID information of the feature vector associated with the feature vector. If the unstructured data is an image, the multimodal anchor table can also store the coordinates of the region of interest {x: 100, y: 200, w: 500, h: 50}; if the unstructured data is audio / video, the multimodal anchor table can also store the timestamp interval {start: 15.5s, end: 20.0s}.

[0078] Step S307: Associate the data table with other data tables and educational management statistical indicators in the pre-built initial graph database to obtain the target graph database.

[0079] Educational management statistical indicators are quantifiable, calculable, and comparable numerical indicators set during the educational management process to describe, analyze, and evaluate the operational status and development level of the education system. These indicators are typically derived from various data within the education management system, such as student information, academic performance data, teaching resources, and teacher data, through statistical analysis and aggregation methods. They are used to support management activities such as educational decision-making, teaching quality assessment, and resource allocation optimization. As a concrete example, educational management statistical indicators may include, but are not limited to, student pass rates and average student grades for certain courses, the number of full-time teachers, and the number of research achievements.

[0080] Furthermore, as mentioned above, different education management systems may have different definitions for the same education management statistical indicator. Therefore, in the graph database of this application, the node information of each node corresponding to the education management statistical indicator may contain not only the calculation formula of the education management statistical indicator, but also the calculation formula in which education management system the calculation formula belongs.

[0081] The initial graph database here can store other data tables and fields from other data tables, such as... Figure 4 The relationships between field 1 in the other data tables shown can also store other data tables and educational management statistical indicators, such as... Figure 4 The relationships between the educational management statistical indicators shown in Figure 2 can also be stored, such as the relationships between educational management statistical indicators. Figure 4 The relationship between education management statistical indicator 1 and education management statistical indicator 2 is shown.

[0082] After obtaining the data table, it can be associated with other data tables and education management statistical indicators in the initial graph database. Subsequently, users can start with the education management statistical indicators and then use the graph database to associate the corresponding data tables.

[0083] Specifically, step S307 includes: Step S3071: Update the nodes corresponding to the data table to the initial graph database to obtain the candidate graph database.

[0084] like Figure 4 As shown, the nodes corresponding to the data tables can be generated first, and then the nodes corresponding to the data tables can be updated in the initial graph database. At this time, there is no relationship between the nodes corresponding to the data tables and other data tables, Education Management Statistical Indicator 1, and Education Management Statistical Indicator 2 in the initial graph database. The node information for the nodes corresponding to the data tables can be the data table name or the second identifier corresponding to the data table.

[0085] Step S3072: Based on the foreign key information corresponding to the data table, construct the first association relationship between the node corresponding to the data table and the node corresponding to other data tables in the candidate graph database, and based on the query statement corresponding to the education management statistical indicators, construct the second association relationship between the node corresponding to the education management statistical indicators and the node corresponding to the data table in the candidate graph database, thus obtaining the target graph database.

[0086] Foreign key information can be fields in a data table used to represent relationships with other data tables, along with their constraint rules. Here, we can first determine other data tables that are related to the data table and the nodes corresponding to those other related data tables based on the foreign key information. Then, in the candidate graph database, we can construct the first association between the nodes corresponding to the data table and the nodes corresponding to the other related data tables.

[0087] A query statement can be a structured query instruction that can be directly executed in the database to calculate a specific educational management statistical indicator. For example, the query statement can be a Structured Query Language (SQL) statement. Since the query statement contains information related to the data table, a second association relationship can be constructed between the nodes corresponding to the educational management statistical indicator and the corresponding nodes in the data table in the candidate graph database based on the query statement.

[0088] The query statement corresponding to the educational management statistical indicators here can be a node in the graph database. The query statement can connect the educational management statistical indicators to data tables (including the data tables mentioned above and others). Of course, the query statement corresponding to the educational management statistical indicators can also be part of the node information corresponding to the educational management statistical indicators, and the node information of the educational management statistical indicators can also include the calculation formulas for the educational management statistical indicators.

[0089] After establishing the first and second associations in the candidate graph database, relationships between data tables and other data tables, as well as relationships between data tables and educational management statistical indicators, can be established. For example, ... Figure 4 As shown, the data table is linked to other data tables, and the data table is also linked to education management statistical indicator 1.

[0090] Based on foreign key information, the first association between data tables and other data tables can be accurately established in the candidate graph database. Furthermore, based on the query statements corresponding to education management statistical indicators, the second association between data tables and education management statistical indicators can be accurately established in the candidate graph database. This allows information originally scattered across different databases and data tables to be uniformly expressed in graph form, thus presenting the inherent connections between data more intuitively. Subsequently, users can start from the nodes corresponding to education management statistical indicators, connect them to the corresponding data tables through the graph database, and query the second metadata corresponding to the data tables from the updated target relational database based on the second identifier of the data tables. Then, based on the second metadata, users can locate the data in the data tables within the education management system and calculate the specific values ​​corresponding to the education statistical indicators.

[0091] Step S308: Based on the second identifier, associate the second metadata in the updated target relational database with the data tables in the target graph database to obtain the updated education index database.

[0092] The specific method for associating the updated second metadata in the target relational database with the data tables in the target graph database based on the second identifier can be referred to the specific method for associating the feature vectors in the target vector library with the first metadata in the target relational database based on the first identifier, as shown above. It will not be repeated here.

[0093] In some alternative implementations, the method further includes: Step c1: Add the nodes corresponding to each field in the second metadata to the target graph database, and establish the association between the nodes corresponding to each field in the target graph database and the nodes corresponding to the data table to obtain the updated target graph database.

[0094] Step c2: Determine the semantic similarity between the first target node and the second target node in the updated target graph database. The first target node is the node corresponding to any field in the second metadata, and the second target node is the node corresponding to any field in other data tables.

[0095] Step c3: If the semantic similarity is greater than or equal to the preset similarity threshold, the first target node and the second target node are merged into the same target node, and the node information of the target node is configured as the fields corresponding to the first target node and the fields corresponding to the second target node.

[0096] like Figure 4As shown, the data table is set to include two fields. First, the nodes corresponding to field 1 and field 2 in the data table can be added to the target graph database. At this time, the nodes corresponding to field 1 and field 2 in the data table are isolated nodes in the target graph database. Therefore, based on the fields and the data tables to which they belong, field 1 and field 2 in the data table can be associated with the data table.

[0097] Furthermore, in the field of educational management, the same concept often has different expressions in different educational management systems. For example, a student academic warning might be called an academic risk or an academic alert. Therefore, fields in data tables from different educational management systems in the target graph database may have the same concept but different expressions. Based on this, after the fields in the data tables are accessed into the target graph database, the semantic similarity between the nodes corresponding to the fields in the data tables (i.e., the first target node) and the nodes corresponding to the existing fields in the target graph database (i.e., the second target node) can be calculated. This semantic similarity is used to determine whether the field information corresponding to the first target node and the field information corresponding to the second target node represent the same concept. If the field information corresponding to the first target node and the field information corresponding to the second target node represent the same concept, then the first target node and the second target node can be merged into a single target node, which can be used to simultaneously represent the field information corresponding to both the first and second target nodes.

[0098] like Figure 4 As shown, since the semantic similarity between field 1 in other data tables and field 2 in data tables is greater than the preset similarity threshold, the nodes corresponding to field 1 in other data tables and the nodes corresponding to field 2 in data tables can be merged into the same node. That is, the node can represent field 1 in other data tables or field 2 in data tables, and the information of the node can include both field 1 in other data tables and field 2 in data tables.

[0099] By calculating the semantic similarity between fields added to the target graph database and fields already in the target graph database, highly similar nodes are automatically merged. This unifies heterogeneous fields into a single target node, resolving the issue of heterogeneous field naming in the education management system. It also enhances the field-level association capabilities of the graph database, significantly improving data consistency, retrieval efficiency, and analytical depth. Furthermore, regardless of the user's preferred expression, they will be able to correctly retrieve the specific data corresponding to that field.

[0100] The method for constructing an educational index database provided in this embodiment associates the updated second metadata in the target relational database with the data tables in the target graph database based on the second identifier corresponding to the data tables. This achieves the fusion of the association between the structured storage of second metadata and the graph storage. In this way, the updated educational index database can improve data retrieval, tracing and analysis capabilities, and also reduce the complexity of data management.

[0101] This embodiment provides a method for querying an educational index database, which can be used on a terminal device. The educational index database is constructed based on the method described above. Figure 5 This is a flowchart of a query method for an educational index database according to an embodiment of the present invention, such as... Figure 5 As shown, the process includes the following steps: Step S501: Receive query information.

[0102] As a concrete example, the query information can be information with a specific query intent. For instance, the query information could be: based on data from the Academic Affairs Office, query the number of full-time faculty members. Another example is: query the average grade of students taking the "Data Structures" course in the Computer Science Department, and so on.

[0103] As a specific example, the educational index database construction system of this application can provide a corresponding interactive page, through which users can input query information, and the terminal device can respond to the user's input and obtain the query information input by the user.

[0104] Step S502: Parse the query information to determine the query type corresponding to the query information.

[0105] The specific method for parsing the query information here can be as follows: a pre-trained intent recognition model is used to identify the intent of the query information and obtain the query intent, i.e., the query category, corresponding to the query information. The pre-trained intent recognition model can be trained based on a large language model architecture, or it can be trained based on a machine learning model architecture. Of course, it can also be trained based on multiple model architectures. There is no specific limitation here, as long as it can identify the query intent, i.e., the query type, corresponding to the query information.

[0106] Of course, you can also segment the query information to obtain multiple entity objects, and then match the entity objects with a pre-set type library to determine the query intent, i.e., the query type, corresponding to the query information. The type library can be used to store the mapping relationship between entity objects and query types.

[0107] Step S503: If the query type represents querying unstructured data, then the query vector corresponding to the query information is matched with the feature vector in the target vector library to obtain the target feature vector that matches the query information.

[0108] When it is determined that the query type corresponding to the query information is unstructured data, the query information can be vectorized first to obtain the query vector corresponding to the query information; then the cosine distance or Euclidean distance between the query vector and each feature vector in the target vector library can be calculated; finally, the feature vector corresponding to the largest cosine distance or Euclidean distance is determined as the target feature vector.

[0109] Step S504: Query the target relational database based on the first identifier corresponding to the target feature vector to obtain the first metadata corresponding to the target feature vector.

[0110] Here, by using the target relational database, the first metadata corresponding to the target feature vector can be retrieved, which is equivalent to determining which education management system the target feature vector comes from and its specific storage location in that education management system.

[0111] Step S505: Based on the first metadata corresponding to the target feature vector, retrieve the target unstructured data corresponding to the target feature vector from the education management system to obtain the target unstructured data corresponding to the query information.

[0112] Based on the first metadata corresponding to the target feature vector, the target unstructured data corresponding to the target feature vector is recalled to obtain the query results corresponding to the query information, i.e., the target unstructured data.

[0113] In addition, after obtaining the target unstructured data corresponding to the query information, a correlation test can be performed between the query information and the target unstructured data. If the correlation test result indicates that the correlation between the query information and the target unstructured data is low, the query information can be further clarified through interaction with the user. Then, the retrieval can be performed again based on the clarified query information until the target unstructured data that meets the requirements is determined.

[0114] In some alternative implementations, the method further includes: Step d1: If the query type represents a query for educational management statistical indicators, then the query statistical indicators in the query information are semantically matched with the educational management statistical indicators in the target graph database to determine the target educational management statistical indicators in the target graph database corresponding to the query information.

[0115] Step d2: Query the target data table associated with the target education management statistical indicators from the target graph database.

[0116] Step d3: Based on the second identifier corresponding to the target data table, query the updated target relational database to obtain the second metadata corresponding to the target data table.

[0117] Step d4: Locate the target data table in the education management system based on the second metadata, and process the data in the target data table based on the query statement corresponding to the target education management statistical indicators to obtain the numerical value corresponding to the query information.

[0118] By using the target graph database, the data table corresponding to the query statistical indicators in the query information can be accurately identified. Furthermore, by using the updated target relational database, the secondary metadata such as the education management system and database name corresponding to the data table can be accurately identified. Then, by using the secondary metadata, the data corresponding to the data table in the education management system can be located. Finally, by using the query statement corresponding to the target education management statistical indicators, the value corresponding to the query information can be accurately determined.

[0119] The query method for the educational index database provided in this embodiment obtains the target feature vector by matching the query vector corresponding to the query information with the feature vector in the target vector library. Then, the first metadata of the target feature vector is determined by the target relational database. The first metadata can then be used to accurately recall the target unstructured data corresponding to the query information, thereby realizing cross-modal data query.

[0120] As a specific application embodiment of the present invention, such as Figure 6 The diagram shown illustrates a specific method for constructing and querying an educational index database according to this application, specifically including steps S601 to S607.

[0121] Step S601: For unstructured data obtained from the education management system, first determine the primary metadata and feature vectors of the unstructured data. The feature vectors are obtained by vectorizing the unstructured data. Then, construct a target relational database based on the primary metadata of the unstructured data; and construct a target vector library based on the corresponding feature vectors of the unstructured data.

[0122] Step S602: For the data table obtained from the education management system, first determine the secondary metadata corresponding to the data table; then update the target relational database based on the secondary metadata corresponding to the data table to obtain a new target relational database (i.e., construct a new target relational database based on the secondary metadata corresponding to the data table); and associate the data table with other data tables and education management statistical indicators in the pre-constructed initial graph database to obtain a target graph database (i.e., construct a target graph database based on the data table).

[0123] Step S603: Construct a ternary collaborative education index database based on the target vector library, the updated target relational database, and the target graph database.

[0124] Step S604: Obtain the user's query information and determine the query type corresponding to the query information.

[0125] Step S605: If the query type represents querying unstructured data, then the target unstructured data corresponding to the query information is determined by using the target vector library and the new target relational database. For details, please refer to the previous text.

[0126] Step S606: If the query type represents a query for educational management statistical indicators, then the numerical values ​​corresponding to the query information are determined by using the target graph database and the new target relational database. For details, please refer to the previous text.

[0127] Step S607: Visualize the target unstructured data or the numerical values ​​corresponding to the query information.

[0128] This embodiment also provides an apparatus for constructing an educational index database, which is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the apparatus described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0129] This embodiment provides an apparatus for constructing an educational index database, such as... Figure 7 As shown, it includes: The first acquisition module 710 is used to acquire the unstructured data corresponding to the education management system, as well as the first identifier and first metadata corresponding to the unstructured data.

[0130] The first processing module 720 is used to perform vectorization processing on unstructured data to obtain the feature vectors corresponding to the unstructured data.

[0131] The first storage module 730 is used to store feature vectors into a pre-built initial vector library to obtain a target vector library, and to store first metadata into a pre-built initial relational database to obtain a target relational database.

[0132] The first association module 740 is used to associate the feature vectors in the target vector library with the first metadata in the target relational database based on the first identifier, so as to obtain the education index database.

[0133] In some optional implementations, the device for obtaining the first metadata corresponding to the unstructured data includes: a first fetch module, used to fetch the storage path and file name of the unstructured data in the education management system; and a first generation module, used to obtain the first metadata corresponding to the unstructured data by fusing the first identifier, file name, and storage path corresponding to the unstructured data.

[0134] In some alternative embodiments, the construction apparatus further includes: The second acquisition module is used to acquire the data table corresponding to the education management system, as well as the second metadata and second identifier of the data table.

[0135] The second storage module is used to store the second metadata into the target relational database, resulting in an updated target relational database.

[0136] The second association module is used to associate the data table with other data tables and educational management statistical indicators in the pre-built initial graph database to obtain the target graph database.

[0137] The third association module is used to associate the updated second metadata in the target relational database with the data tables in the target graph database based on the second identifier, so as to obtain the updated education index database.

[0138] In some optional implementations, the device for obtaining the second metadata corresponding to the data table includes: a second retrieval module, used to retrieve the database name, data table name, and field names of the data table within the education management system; and a second generation module, used to obtain the second metadata based on the fusion of the second identifier, database name, data table name, and field names.

[0139] In some optional implementations, the second association module is further configured to update the nodes corresponding to the data table to the initial graph database to obtain a candidate graph database; based on the foreign key information corresponding to the data table, construct a first association relationship between the nodes corresponding to the data table and the nodes corresponding to other data tables; construct the first association relationship between the nodes corresponding to the data table and the nodes corresponding to other data tables in the candidate graph database, and based on the query statement corresponding to the education management statistical indicators, construct a second association relationship between the nodes corresponding to the education management statistical indicators and the nodes corresponding to the data tables in the candidate graph database to obtain a target graph database.

[0140] In some optional implementations, the construction apparatus further includes: an update module, configured to add nodes corresponding to each field in the second metadata to the target graph database, and establish associations between nodes corresponding to each field in the target graph database and nodes corresponding to the data tables, thereby obtaining an updated target graph database; a first determination module, configured to determine the semantic similarity between a first target node and a second target node in the updated target graph database, wherein the first target node is a node corresponding to any field in the second metadata, and the second target node is a node corresponding to any field in other data tables; and a merging module, configured to merge the first target node and the second target node into a single target node if the semantic similarity is greater than or equal to a preset similarity threshold, wherein the node information of the target node is configured to include the field information corresponding to the first target node and the field information corresponding to the second target node.

[0141] This embodiment provides a query device for an educational index database. The educational index database is constructed based on the device described above, such as... Figure 8 As shown, it includes: The second acquisition module 810 is used to receive query information.

[0142] The parsing module 820 is used to parse the query information and determine the query type corresponding to the query information.

[0143] The first query module 830 is used to perform vector matching between the query vector corresponding to the query information and the feature vector in the target vector library if the query type represents unstructured data, so as to obtain the target feature vector that matches the query information.

[0144] The second query module 840 is used to query the target relational database based on the first identifier corresponding to the target feature vector to obtain the first metadata corresponding to the target feature vector.

[0145] The third retrieval module 850 is used to retrieve the target unstructured data corresponding to the target feature vector from the education management system based on the first metadata corresponding to the target feature vector, and obtain the target unstructured data corresponding to the query information.

[0146] In some optional implementations, the construction apparatus further includes: a second determining module, configured to semantically match the statistical indicators in the query information with the educational management statistical indicators in the target graph database if the query type represents a query statistical indicator, to determine the target educational management statistical indicator in the target graph database corresponding to the query information; a third query module, configured to query the target data table associated with the target educational management statistical indicator from the target graph database; a fourth query module, configured to query the updated target relational database based on the second identifier corresponding to the target data table to obtain the second metadata corresponding to the target data table; and a second processing module, configured to retrieve the target data table from the educational management system based on the second metadata, and process the data in the target data table based on the query statement corresponding to the target educational management statistical indicator to obtain the value corresponding to the query information.

[0147] The educational index database construction apparatus provided in this embodiment of the invention can execute the educational index database construction method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method execution. The educational index data query apparatus provided in this embodiment of the invention can execute the educational index database query method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method execution. Further functional descriptions of the above modules and units are the same as in the corresponding embodiments described above, and will not be repeated here.

[0148] Figure 9 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. The electronic device may be the terminal device or server described above.

[0149] The following is a detailed reference. Figure 9 The diagram illustrates a structural schematic suitable for implementing an electronic device according to embodiments of the present invention. The electronic device may include a processor (e.g., a central processing unit, graphics processor, etc.) 901, which can perform various appropriate actions and processes based on a program stored in read-only memory (ROM) 902 or a program loaded from memory 908 into random access memory (RAM) 903. The random access memory 903 also stores various programs and data required for the operation of the electronic device. The processor 901, ROM 902, and RAM 903 are interconnected via a bus 904. An input / output (I / O) interface 905 is also connected to the bus 904.

[0150] Typically, the following devices can be connected to the input / output interface 905: input devices 906 including, for example, a touchscreen, touchpad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; output devices 907 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; memory devices 908 including, for example, magnetic tape, hard disk, etc.; and communication devices 909. Communication devices 909 allow electronic devices to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 9 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown, and more or fewer devices may be implemented or have instead.

[0151] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 909, or installed from a memory 908, or installed from a read-only memory 902. When the computer program is executed by the processor 901, it performs the functions defined in the method for constructing an educational index database according to embodiments of the present invention.

[0152] Figure 9 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0153] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the method for constructing the educational index database shown in the above embodiments is implemented.

[0154] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0155] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for constructing an educational index database, characterized in that, The method includes: Obtain the unstructured data corresponding to the education management system, as well as the first identifier and first metadata corresponding to the unstructured data; The unstructured data is vectorized to obtain the feature vector corresponding to the unstructured data; The feature vectors are stored in a pre-built initial vector library to obtain a target vector library, and the first metadata is stored in a pre-built initial relational database to obtain a target relational database; Based on the first identifier, the feature vector in the target vector library and the first metadata in the target relational database are associated to obtain an education index database.

2. The method for constructing an educational index database according to claim 1, characterized in that, The process of obtaining the first metadata corresponding to the unstructured data includes: Retrieve the storage path and file name of the unstructured data in the education management system; The first metadata corresponding to the unstructured data is obtained by fusing the first identifier, file name, and storage path corresponding to the unstructured data.

3. The method for constructing an educational index database according to claim 1, characterized in that, The method further includes: Obtain the data table corresponding to the education management system, as well as the second metadata and second identifier corresponding to the data table; The second metadata is stored in the target relational database to obtain the updated target relational database; The data tables are associated with other data tables and educational management statistical indicators in the pre-constructed initial graph database to obtain the target graph database; Based on the second identifier, the second metadata in the updated target relational database and the data table in the target graph database are associated to obtain the updated education index database.

4. The method for constructing an educational index database according to claim 3, characterized in that, The process of obtaining the second metadata corresponding to the data table includes: Retrieve the database name, table name, and field names of the data table within the education management system; The second metadata is obtained by combining the second identifier, the database name, the data table name, and the names of each field.

5. The method for constructing an educational index database according to claim 3, characterized in that, The step of associating the data table with other data tables and educational management statistical indicators in the pre-constructed initial graph database to obtain the target graph database includes: Update the nodes corresponding to the data table to the initial graph database to obtain the candidate graph database; Based on the foreign key information corresponding to the data table, a first association relationship is constructed between the node corresponding to the data table and the node corresponding to the other data tables in the candidate graph database. Based on the query statement corresponding to the education management statistical indicator, a second association relationship is constructed between the node corresponding to the education management statistical indicator and the node corresponding to the data table in the candidate graph database, thus obtaining the target graph database.

6. The method for constructing an educational index database according to claim 4, characterized in that, The method further includes: Add the nodes corresponding to each field in the second metadata to the target graph database, and establish the association relationship between the nodes corresponding to each field and the nodes corresponding to the data table in the target graph database to obtain the updated target graph database; Determine the semantic similarity between the first target node and the second target node in the updated target graph database, wherein the first target node is the node corresponding to any field in the second metadata, and the second target node is the node corresponding to any field in the other data tables; If the semantic similarity is greater than or equal to a preset similarity threshold, the first target node and the second target node are merged into the same target node, and the node information of the target node is configured as the field information corresponding to the first target node and the field information corresponding to the second target node.

7. A method for querying an educational index database, wherein the educational index database is constructed according to any one of claims 1 to 6, characterized in that, The method further includes: Receive query information; The query information is parsed to determine the query type corresponding to the query information; If the query type represents querying unstructured data, then the query vector corresponding to the query information is matched with the feature vector in the target vector library to obtain the target feature vector that matches the query information. Based on the first identifier corresponding to the target feature vector, query the target relational database to obtain the first metadata corresponding to the target feature vector; Based on the first metadata corresponding to the target feature vector, the target unstructured data corresponding to the target feature vector is retrieved from the education management system to obtain the target unstructured data corresponding to the query information.

8. The query method for the educational index database according to claim 7, characterized in that, The method further includes: If the query type represents a query for educational management statistical indicators, then the query statistical indicators in the query information are semantically matched with the educational management statistical indicators in the target graph database to determine the target educational management statistical indicators in the target graph database corresponding to the query information. Query the target data table associated with the target education management statistical indicators from the target graph database; Based on the second identifier corresponding to the target data table, query the updated target relational database to obtain the second metadata corresponding to the target data table; Based on the second metadata, the target data table in the education management system is retrieved, and the data in the target data table is processed based on the query statement corresponding to the target education management statistical indicator to obtain the value corresponding to the query information.

9. An apparatus for constructing an educational index database, characterized in that, The device includes: The first acquisition module is used to acquire unstructured data corresponding to the education management system, as well as the first identifier and first metadata corresponding to the unstructured data; The first processing module is used to perform vectorization processing on the unstructured data to obtain the feature vector corresponding to the unstructured data; The first storage module is used to store the feature vectors into a pre-built initial vector library to obtain a target vector library, and to store the first metadata into a pre-built initial relational database to obtain a target relational database; The first association module is used to associate the feature vector in the target vector library with the first metadata in the target relational database based on the first identifier, so as to obtain an education index database.

10. An electronic device, characterized in that, include: The system includes a memory and a processor, which are communicatively connected to each other. The memory stores computer instructions, and the processor executes the computer instructions to perform either the method for constructing the educational index database according to any one of claims 1 to 6, or the method for querying the educational index database according to any one of claims 7 to 8.

Images