Data query method and device for data islands, equipment and medium

Abstract

The invention relates to the technical field of data processing, and discloses a data query method and device for a data island, equipment and a medium, and the method comprises the steps: carrying out the data integration of a data set in the data island, and obtaining an integrated data set; calculating the number of features of each piece of integrated data in the integrated data set, and constructing a multi-channel feature map according to the number of features; calculating target data features according to the multi-channel feature map, and performing data classification on the integrated data set according to the target data features to obtain a classified data set; determining associated data of each piece of integrated data according to the classification data set, and creating an associated data index according to the associated data; and when a data query request is received, analyzing target data of the data query request according to the associated data index. According to the invention, the accuracy of data query in the data island can be improved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a data query method, device, equipment and medium for data islands. Background Art

[0002] With the development of the big data era, data electronicization has become an office trend, but different companies, departments, and even individuals have different ways of expressing data, so that when the amount of data is large, it is impossible to achieve data intercommunication, that is, there are data islands between data, resulting in the inability to accurately query data, and many situations such as missed checks and wrong checks occur.

[0003] However, the existing methods for data islands mainly include building data warehouses based on Hadoop ecological components, that is, extracting and converting the original data stored in each data island, building an ODS layer (Operational Data Store), realizing data integration, and querying the integrated data. However, the data will be stored repeatedly when the operational data store is stored, resulting in poor accuracy in subsequent data queries. Therefore, how to improve the accuracy of data island data queries has become a problem that needs to be solved urgently. Summary of the invention

[0004] The present invention provides a data query method, device, equipment and medium for data islands, the main purpose of which is to solve the problem of poor accuracy of data query in data islands.

[0005] To achieve the above object, the present invention provides a data query method for data islands, comprising:

[0006] Acquire a data set in a data island, and perform data integration on the data set to obtain an integrated data set;

[0007] Calculating the number of features of each integrated data in the integrated data set, and constructing a multi-channel feature map of the integrated data according to the number of features;

[0008] Calculate the target data feature of each integrated data according to the multi-channel feature map, and classify the integrated data set according to the target data feature to obtain a classified data set;

[0009] Determine the associated data of each of the integrated data according to the classified data set, and create an associated data index according to the associated data;

[0010] When a data query request is received, target data of the data query request is parsed according to the associated data index.

[0011] Optionally, performing data integration on the data set to obtain an integrated data set includes:

[0012] Classifying the data set according to data format to obtain a structured data set and an unstructured data set;

[0013] Performing data standardization and data deduplication processing on the structured data set to obtain a target structured data set;

[0014] Performing data verification and data denoising on the unstructured data set to obtain a target unstructured data set;

[0015] The target structured data and the target unstructured data set are collected to obtain an integrated data set.

[0016] Optionally, calculating the number of features of each integrated data in the integrated data set includes:

[0017] identifying the data type of each of the integrated data;

[0018] Performing feature conversion on each of the integrated data according to the data type to obtain converted data;

[0019] The number of features of each of the integrated data is calculated according to the transformed data.

[0020] Optionally, constructing a multi-channel feature map of the integrated data according to the number of features includes:

[0021] Extracting characteristic data values ​​of the integrated data, and constructing a characteristic array of the characteristic data values ​​according to the characteristic quantity;

[0022] Generate multiple element arrays according to the feature data, and splice the element arrays to obtain multiple spliced ​​arrays;

[0023] The spliced ​​arrays are superimposed to obtain a multi-channel feature map of the integrated data.

[0024] Optionally, calculating the target data feature of each integrated data according to the multi-channel feature map includes:

[0025] Performing squeezing excitation on the multi-channel feature map to obtain a convolution excitation feature;

[0026] The multi-channel feature map is squeezed and excited using the following formula:

[0027] x′=sigmoid(FC2(ReLU(FC1(GAP(x)))))

[0028] Wherein, x′ represents the convolutional excitation feature, sigmoid represents the sigmoid activation function, FC2 represents the preset second fully connected layer, ReLU represents the ReLU activation function, FC1 represents the preset first fully connected layer, GAP represents the global pooling operation, and x represents the multi-channel feature map;

[0029] Performing feature segmentation on the convolution excitation feature to obtain a segmentation feature set, and performing convolution on each segmentation feature in the segmentation feature set to obtain a segmentation convolution feature;

[0030] Feature concatenation is performed on the segmented convolution features to obtain target data features of each of the integrated data.

[0031] Optionally, the creating an associated data index according to the associated data includes:

[0032] Identify, according to the associated data, associated island data of each data in the data island and an associated category corresponding to the associated island data;

[0033] Determine an index keyword of the associated data according to the associated island data and the associated category;

[0034] An associated data index for each data in the data island is constructed according to the index keyword.

[0035] Optionally, parsing the target data of the data query request according to the associated data index includes:

[0036] Obtain a request message corresponding to the data query request, strip and parse the request message, and obtain application layer data of the request message;

[0037] Identifying the message type of the request message according to the application layer data, and invoking a pre-built parsing rule according to the message type to parse the request target of the request message;

[0038] According to the request target, data is extracted from a preset database using the associated data index to obtain target data corresponding to the data query request.

[0039] In order to solve the above problem, the present invention further provides a data query device for data islands, the device comprising:

[0040] A data integration module is used to obtain data sets in data islands, perform data integration on the data sets, and obtain integrated data sets;

[0041] A multi-channel feature map construction module, used to calculate the number of features of each integrated data in the integrated data set, and construct a multi-channel feature map of the integrated data according to the number of features;

[0042] A data classification module, used for calculating the target data feature of each of the integrated data according to the multi-channel feature map, and classifying the integrated data set according to the target data feature to obtain a classified data set;

[0043] A related data index creation module, used to determine the related data of each integrated data according to the classified data set, and create a related data index according to the related data;

[0044] The data query module is used to parse the target data of the data query request according to the associated data index when receiving the data query request.

[0045] In order to solve the above problem, the present invention further provides an electronic device, the electronic device comprising:

[0046] at least one processor; and,

[0047] a memory communicatively connected to the at least one processor; wherein,

[0048] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the data query method for data islands described above.

[0049] In order to solve the above problem, the present invention also provides a computer-readable storage medium, in which at least one computer program is stored. The at least one computer program is executed by a processor in an electronic device to implement the above-mentioned data query method for data islands.

[0050] The embodiment of the invention can improve data quality by integrating the data sets in the data islands, and realize the integration and unified management of the data island data sets, which is convenient for subsequent data query; construct a multi-channel feature map for each integrated data, calculate the target data features according to the multi-channel feature map, and accurately classify the integrated data sets using the feature diversity of the target data features to obtain a classified data set; determine the associated data of each integrated data according to the classified data set, and create an associated data index; when receiving a data query request, the target data of the data query request can be parsed according to the associated data index, so as to realize data intercommunication between databases forming the data islands, avoid missed queries and wrong queries, and improve the accuracy of data island data queries. Therefore, the data query method, device, equipment and medium for data islands proposed by the present invention can solve the problem of poor accuracy of data island data queries. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1A schematic diagram of a flow chart of a data query method for data islands provided by an embodiment of the present invention;

[0052] Figure 2 A schematic diagram of a process of integrating a data set to obtain an integrated data set provided by an embodiment of the present invention;

[0053] Figure 3 A schematic diagram of a process for constructing a multi-channel feature map of integrated data provided by an embodiment of the present invention;

[0054] Figure 4 A functional module diagram of a data query device for data islands provided by an embodiment of the present invention;

[0055] Figure 5 A schematic diagram of the structure of an electronic device for implementing the data query method for data islands provided in one embodiment of the present invention.

[0056] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0057] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0058] The embodiment of the present application provides a data query method for data islands. The execution subject of the data query method for data islands includes but is not limited to at least one of the electronic devices such as a server, a terminal, etc. that can be configured to execute the method provided by the embodiment of the present application. In other words, the data query method for data islands can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server, or it can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and big data and artificial intelligence platforms.

[0059] Reference Figure 1 FIG. 1 is a flow chart of a data query method for data islands provided by an embodiment of the present invention. In this embodiment, the data query method for data islands includes:

[0060] S1. Obtain a data set in a data island, and perform data integration on the data set to obtain an integrated data set.

[0061] In an embodiment of the present invention, a data island may be a form of data isolation formed by different operating systems and different types of databases. The data set includes data in the database corresponding to the data island. Data integration is to combine the data in different data islands so as to provide an organization with a unified view for the data island and realize unified management and analysis of data. Among them, the data set in the data island may be supply chain data in the financial field. The data between different supply chain links may form data islands, which may lead to.

[0062] In the embodiment of the present invention, refer to Figure 2 As shown, the data integration is performed on the data set to obtain an integrated data set, including:

[0063] S21, classifying the data set into data formats to obtain a structured data set and an unstructured data set;

[0064] S22, performing data standardization and data deduplication processing on the structured data set to obtain a target structured data set;

[0065] S23, performing data verification and data denoising processing on the unstructured data set to obtain a target unstructured data set;

[0066] S24: Collect the target structured data and the target unstructured data set to obtain an integrated data set.

[0067] In an embodiment of the present invention, the structured data set includes structured data in the data set, such as table data and other data with a fixed format and pattern, and the unstructured data set includes data without a fixed format or pattern, such as text, pictures, videos, and audio data.

[0068] Preferably, the unstructured data may also include semi-structured data such as XML files and JSON files. The semi-structured data may be parsed to identify fields or tags therein, and data verification and data noise removal may be performed on the semi-structured data.

[0069] In an embodiment of the present invention, data standardization for structured data is to convert the format of structured data and the unit of data into a unified format, such as format conversion of data such as date and time. Further, deduplication can be performed using a hash deduplication method. Data verification for unstructured data can be to verify the integrity of the data, whether the file format of the data complies with its declared format, the range of the data, and the consistency of the data, etc., and noise removal can be performed using median filtering or Gaussian filtering.

[0070] In the embodiment of the present invention, by integrating the data set, the data quality in the integrated data set can be improved, and at the same time, the data of different formats in the data island can be combined, so that the integration and unified management of the data island data set can be realized, which is convenient for subsequent data query.

[0071] S2. Calculate the number of features of each integrated data in the integrated data set, and construct a multi-channel feature map of the integrated data according to the number of features.

[0072] In an embodiment of the present invention, the integrated data set includes integrated data of different formats and categories. The number of features can be calculated based on the number of variables or attributes that can be used for analysis of each integrated data. For example, in a table, each column usually represents a feature, and attributes such as age and price represented by data values ​​can represent a feature.

[0073] In the embodiment of the present invention, the step of calculating the number of features of each integrated data in the integrated data set includes:

[0074] identifying the data type of each of the integrated data;

[0075] Performing feature conversion on each of the integrated data according to the data type to obtain converted data;

[0076] The number of features of each of the integrated data is calculated according to the transformed data.

[0077] In an embodiment of the present invention, different types of integrated data are used for different data features. For example, the number of features of image data is usually related to the resolution and color channels of the image, and the number of features of time series data may include timestamps and measurement values ​​associated therewith. Therefore, it is necessary to calculate the number of features based on different data types. Specifically, a pre-built programming language can be used to calculate the number of features of each integrator based on the data type of the integrated data.

[0078] Specifically, feature conversion is to convert the integrated data into data that can directly reflect the number of features according to the type. For example, for numerical data, the numerical variable is usually regarded as a separate feature, and the feature data of the numerical data is calculated by identifying the numerical variable. For text data, the text can be converted into numerical features. After the conversion, the number of features depends on the number of different words. For categorical data, One-Hot Encoding can be used to convert the categorical data into multiple binary features to calculate the number of features.

[0079] In an embodiment of the present invention, a multi-channel feature map converts the data features of the integrated data corresponding to the number of features from one-dimensional structured unbalanced data to multi-channel unbalanced data in the form of a three-dimensional array, providing more possible feature values ​​for subsequent calculation of target data features and improving the accuracy of subsequent target data feature calculations.

[0080] In the embodiment of the present invention, refer to Figure 3 As shown, constructing the multi-channel feature map of the integrated data according to the number of features includes:

[0081] S31, extracting characteristic data values ​​of the integrated data, and constructing a characteristic array of the characteristic data values ​​according to the characteristic quantity;

[0082] S32, generating a plurality of element arrays according to the characteristic data, and splicing the element arrays to obtain a plurality of spliced ​​arrays;

[0083] S33, superimposing the spliced ​​arrays to obtain a multi-channel feature map of the integrated data.

[0084] In an embodiment of the present invention, the data feature value may be a value corresponding to each data feature in the integrated data, such as the resolution of the image data, the value of each data attribute, etc. By combining the feature data values, a one-dimensional feature array composed of the feature data values ​​is obtained, wherein the size of the feature data is consistent with the number of features. Specifically, non-numeric data features may be filled with 0, such as label features, to obtain a one-dimensional feature array composed of the feature data values.

[0085] Specifically, the feature data values ​​in the feature array are randomly shuffled and combined into n two-dimensional K×K element arrays, the element arrays are horizontally spliced ​​to obtain array A, the steps of calculating array A are repeated to obtain n arrays A, the n arrays A are vertically spliced ​​to obtain a spliced ​​array, the steps of calculating the spliced ​​array are repeated, for example, x times to obtain x spliced ​​arrays, the spliced ​​arrays are superimposed to obtain a three-dimensional multi-channel feature map. After the x two-dimensional spliced ​​arrays are superimposed, a three-dimensional multi-channel feature map can be obtained.

[0086] In the embodiment of the present invention, by constructing a multi-channel feature map and randomly combining adjacent feature data values ​​in the integrated data, the diversity of subsequent target data feature extraction can be improved to obtain target data features with more accurate feature information.

[0087] S3. Calculate the target data features of each integrated data according to the multi-channel feature map, and classify the integrated data set according to the target data features to obtain a classified data set.

[0088] In the embodiment of the present invention, the target data feature is a feature vector constructed according to the multi-channel feature map, and the target data feature can be used to deeply mine the feature information in the multi-channel feature map.

[0089] In the embodiment of the present invention, the step of calculating the target data feature of each integrated data according to the multi-channel feature map includes:

[0090] Performing squeezing excitation on the multi-channel feature map to obtain a convolution excitation feature;

[0091] Performing feature segmentation on the convolution excitation feature to obtain a segmentation feature set, and performing convolution on each segmentation feature in the segmentation feature set to obtain a segmentation convolution feature;

[0092] Feature concatenation is performed on the segmented convolution features to obtain target data features of each of the integrated data.

[0093] In an embodiment of the present invention, squeeze excitation can calculate the importance of convolution channel information in a multi-channel feature map, thereby obtaining a more accurate convolution excitation feature, and slicing is performed according to the feature dimension of the convolution excitation feature. For example, the channel dimension of the convolution excitation feature is 8×8×128, and the channel dimension 16 is used as the segmentation length. The convolution excitation feature can be segmented into 8 segmentation features of 8×8×16 dimensions to obtain a segmentation feature set. The segmentation feature set can be used to more comprehensively extract feature information in the multi-channel feature map, thereby obtaining a more accurate target data feature.

[0094] Specifically, the multi-channel feature map is squeezed and excited using the following formula:

[0095] x′=sigmoid(FC2(ReLU(FC1(GAP(x)))))

[0096] Among them, x′ represents the convolution excitation feature, sigmoid represents the sigmoid activation function, FC2 represents the preset second fully connected layer, ReLU represents the ReLU activation function, FC1 represents the preset first fully connected layer, GAP represents the global pooling operation, and x represents the multi-channel feature map.

[0097] In an embodiment of the present invention, the integrated data set can be classified by connecting the fully connected layer after feature splicing to obtain classified data sets of different categories. For example, the classified data set can be obtained by classification according to data source, business type, time, business department, etc.

[0098] S4. Determine associated data of each integrated data according to the classified data set, and create an associated data index according to the associated data.

[0099] In the embodiment of the present invention, associated data refers to two different integrated data that can be connected to each other through a certain relationship, for example, having the same data source, the same business category, etc. Therefore, the associated data of each integrated data can be determined by the classified data set. For example, if integrated data 1 and integrated data 2 are in the classified data set composed of the same business category, then integrated data 1 and integrated data 2 are associated data with the same business category.

[0100] In the embodiment of the present invention, the associated data index is a data structure that allows a database or a data processing system to quickly locate the physical location of data corresponding to the associated data index, thereby speeding up data query.

[0101] Specifically, the creating an associated data index according to the associated data includes:

[0102] Identify, according to the associated data, associated island data of each data in the data island and an associated category corresponding to the associated island data;

[0103] Determine an index keyword of the associated data according to the associated island data and the associated category;

[0104] An associated data index for each data in the data island is constructed according to the index keyword.

[0105] In an embodiment of the present invention, the associated island data of each data can be identified according to the corresponding data after the associated data is integrated in the data island. For example, associated data 1 is the data after data a in the data island is integrated, and the associated island data of data a in the data island is associated data 1; the associated category is the association relationship of the associated data, such as business category, data source, etc.

[0106] Furthermore, the index keyword can be the physical location and associated category corresponding to the associated island data of each data, wherein the physical location is the location where the associated island data is stored. The physical location and the associated category are used as index keywords for the associated data, and the associated data index is created using the CREATE INDEX statement or the ALTER TABLE statement using the index keyword as the index name.

[0107] In an embodiment of the present invention, the associated data of each data can be directly pointed to through the associated data index, and the association relationship between the associated data can be determined through the associated data index, so that diversified data queries can be implemented from different data storages to prevent data omissions and mis-checks, thereby improving the accuracy of data queries.

[0108] S5. When a data query request is received, the target data of the data query request is parsed according to the associated data index.

[0109] In an embodiment of the present invention, a data query request is a request for data in a data island to obtain specific information or data sets. The data query request can be a simple query for a specific data or a complex one, for example, involving the connection, filtering, sorting and aggregation of multiple data. Therefore, it is necessary to parse the target data of the data query request according to the associated data index to achieve intercommunication between data island data and improve the accuracy of data query.

[0110] In the embodiment of the present invention, the step of parsing the target data of the data query request according to the associated data index includes:

[0111] Obtain a request message corresponding to the data query request, strip and parse the request message, and obtain application layer data of the request message;

[0112] Identifying the message type of the request message according to the application layer data, and invoking a pre-built parsing rule according to the message type to parse the request target of the request message;

[0113] According to the request target, data is extracted from a preset database using the associated data index to obtain target data corresponding to the data query request.

[0114] In an embodiment of the present invention, a request message is a message sent by a client to a server to request target data. The request message includes a request line and a request body. The HTTP method (such as GET, POST, PUT, DELETE, etc.) is extracted from the request line, and the resource path (URI) in the request line is parsed. The resource path usually indicates the target of the requested resource or operation, that is, the request target can be obtained by parsing the request message, wherein the request target is the data requested by the data query request.

[0115] In the embodiment of the present invention, stripping and parsing the request message can obtain application layer data including the status line, request header, blank line and optional request body in the request message. Specifically, a pre-built analysis engine can be called to strip and parse the request message. After obtaining the application layer data, the message type of the request message can be determined according to the byte data of the specified bytes in the application layer data, and the request message can be parsed using the parsing rules corresponding to the message type to obtain the request target of the request message.

[0116] Among them, the message type may include a login request, a logout request or an SQL request, and the message type can be determined according to the first byte of the application layer data. For example, if the byte data of the first byte in the application layer data is 0x00, the message type can be determined to be a login request; if the byte data of the first byte is 0x58, the message type of the request message can be determined to be a logout request; if the byte data of the first byte is 0x50, the message type of the request message can be determined to be an SQL request, and then the request message can be parsed according to specific message parsing rules to obtain the request target.

[0117] In an embodiment of the present invention, the preset database is a database that forms a data island. Data retrieval is performed in the database through the request data pointed to by the request target and the associated data index of the request data, and the target data corresponding to the request target can be obtained. For example, the request target is data 1 and the associated data of the same business category, and the target data can be extracted from the database using the request target and the associated data pointed to by the associated data index.

[0118] In the embodiment of the present invention, by parsing the target data through the associated data index, data intercommunication between databases forming data islands can be achieved, and the target data of the data query request can be accurately extracted to avoid missed searches and wrong searches, thereby improving the accuracy of data island data queries.

[0119] like Figure 4 , which is a functional module diagram of a data query device for data islands provided by an embodiment of the present invention.

[0120] The data query device 100 for data islands of the present invention can be installed in an electronic device. According to the functions to be implemented, the data query device 100 for data islands can include a data integration module 101, a multi-channel feature map construction module 102, a data classification module 103, an associated data index creation module 104 and a data query module 105. The module of the present invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0121] In this embodiment, the functions of each module / unit are as follows:

[0122] The data integration module 101 is used to obtain the data sets in the data islands, perform data integration on the data sets, and obtain an integrated data set;

[0123] The multi-channel feature map construction module 102 is used to calculate the number of features of each integrated data in the integrated data set, and construct a multi-channel feature map of the integrated data according to the number of features;

[0124] The data classification module 103 is used to calculate the target data feature of each integrated data according to the multi-channel feature map, and classify the integrated data set according to the target data feature to obtain a classified data set;

[0125] The associated data index creation module 104 is used to determine the associated data of each integrated data according to the classified data set, and create an associated data index according to the associated data;

[0126] The data query module 105 is configured to parse the target data of the data query request according to the associated data index when receiving the data query request.

[0127] In detail, each module described in the data query device 100 for data islands in the embodiment of the present invention is used in the same manner as described above. Figures 1 to 3 The data query method for data islands described in the text is the same as the technical means and can produce the same technical effects, so I will not go into details here.

[0128] like Figure 5 , which is a schematic diagram of the structure of an electronic device for implementing a data query method for data islands provided by an embodiment of the present invention.

[0129] The electronic device 1 may include a processor 10, a memory 11, a communication bus 12, and a communication interface 13, and may also include a computer program stored in the memory 11 and executable on the processor 10, such as a data query program for data islands.

[0130] Among them, the processor 10 can be composed of an integrated circuit in some embodiments. The processor 10 is the control core (Control Unit) of the electronic device, and uses various interfaces and lines to connect various components of the entire electronic device. It runs or executes programs or modules stored in the memory 11 (for example, executing data query programs for data islands, etc.), and calls data stored in the memory 11 to execute various functions of the electronic device and process data.

[0131] The memory 11 includes at least one type of readable storage medium, including flash memory, mobile hard disk, multimedia card, card-type memory (such as SD or DX memory, etc.), magnetic memory, disk, optical disk, etc. The memory 11 can be used not only to store application software installed in the electronic device and various data, such as the code of the data query program for the data island, but also to temporarily store the data that has been output or is to be output.

[0132] The communication bus 12 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. The bus is configured to realize connection and communication between the memory 11 and at least one processor 10, etc.

[0133] The communication interface 13 is used for communication between the electronic device and other devices, and includes a network interface and a user interface.

[0134] The figure only shows an electronic device with components. Those skilled in the art will understand that the structure shown in the figure does not constitute a limitation on the electronic device, and may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0135] For example, although not shown, the electronic device may also include a power source (such as a battery) for supplying power to each component. Preferably, the power source may be logically connected to the at least one processor 10 through a power management device, so that the power management device can realize functions such as charging management, discharging management, and power consumption management. The power source may also include one or more DC or AC power sources, recharging devices, power failure detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The electronic device may also include a variety of sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be repeated here.

[0136] The present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor of an electronic device, the computer program can implement:

[0137] Acquire a data set in a data island, and perform data integration on the data set to obtain an integrated data set;

[0138] Calculating the number of features of each integrated data in the integrated data set, and constructing a multi-channel feature map of the integrated data according to the number of features;

[0139] Calculate the target data feature of each integrated data according to the multi-channel feature map, and classify the integrated data set according to the target data feature to obtain a classified data set;

[0140] Determine the associated data of each of the integrated data according to the classified data set, and create an associated data index according to the associated data;

[0141] When a data query request is received, target data of the data query request is parsed according to the associated data index.

[0142] It should be understood that the embodiment is for illustration only and the scope of the patent application is not limited to this structure.

[0143] Specifically, the specific implementation method of the processor 10 for the above instructions can refer to the description of the relevant steps in the corresponding embodiment of the accompanying drawings, which will not be repeated here.

[0144] Furthermore, if the module / unit integrated in the electronic device 1 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disk, a computer memory, and a read-only memory (ROM).

[0145] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A data query method for data islands, characterized in that: The method comprises: Acquire a data set in a data island, and perform data integration on the data set to obtain an integrated data set; Calculating the number of features of each integrated data in the integrated data set, and constructing a multi-channel feature map of the integrated data according to the number of features; Calculate the target data feature of each integrated data according to the multi-channel feature map, and classify the integrated data set according to the target data feature to obtain a classified data set; Determine the associated data of each of the integrated data according to the classified data set, and create an associated data index according to the associated data; When a data query request is received, target data of the data query request is parsed according to the associated data index.

2. The data query method for data islands according to claim 1, characterized in that: The step of integrating the data set to obtain an integrated data set includes: Classifying the data set according to data format to obtain a structured data set and an unstructured data set; Performing data standardization and data deduplication processing on the structured data set to obtain a target structured data set; Performing data verification and data denoising on the unstructured data set to obtain a target unstructured data set; The target structured data and the target unstructured data set are collected to obtain an integrated data set.

3. The data query method for data islands according to claim 1, characterized in that: The calculating the number of features of each integrated data in the integrated data set comprises: identifying the data type of each of the integrated data; Performing feature conversion on each of the integrated data according to the data type to obtain converted data; The number of features of each of the integrated data is calculated according to the transformed data.

4. The data query method for data islands according to claim 1, characterized in that: The step of constructing a multi-channel feature map of the integrated data according to the number of features comprises: Extracting characteristic data values ​​of the integrated data, and constructing a characteristic array of the characteristic data values ​​according to the characteristic quantity; Generate multiple element arrays according to the feature data, and splice the element arrays to obtain multiple spliced ​​arrays; The spliced ​​arrays are superimposed to obtain a multi-channel feature map of the integrated data.

5. The data query method for data islands according to claim 1, characterized in that: The step of calculating the target data feature of each integrated data according to the multi-channel feature map comprises: Performing squeezing excitation on the multi-channel feature map to obtain a convolution excitation feature; The multi-channel feature map is squeezed and excited using the following formula: x′=sigmoid(FC2(ReLU(FC1(GAP(x))))) Wherein, x′ represents the convolutional excitation feature, sigmoid represents the sigmoid activation function, FC2 represents the preset second fully connected layer, ReLU represents the ReLU activation function, FC1 represents the preset first fully connected layer, GAP represents the global pooling operation, and x represents the multi-channel feature map; Performing feature segmentation on the convolution excitation feature to obtain a segmentation feature set, and performing convolution on each segmentation feature in the segmentation feature set to obtain a segmentation convolution feature; Feature concatenation is performed on the segmented convolution features to obtain target data features of each of the integrated data.

6. The data query method for data islands according to claim 1, characterized in that: The step of creating an associated data index according to the associated data includes: Identify, according to the associated data, associated island data of each data in the data island and an associated category corresponding to the associated island data; Determine an index keyword of the associated data according to the associated island data and the associated category; An associated data index for each data in the data island is constructed according to the index keyword.

7. The data query method for data islands according to claim 1, characterized in that: The step of parsing the target data of the data query request according to the associated data index includes: Obtain a request message corresponding to the data query request, strip and parse the request message, and obtain application layer data of the request message; Identifying the message type of the request message according to the application layer data, and invoking a pre-built parsing rule according to the message type to parse the request target of the request message; According to the request target, data is extracted from a preset database using the associated data index to obtain target data corresponding to the data query request.

8. A data query device for data islands, characterized in that: The device comprises: A data integration module is used to obtain data sets in data islands, perform data integration on the data sets, and obtain integrated data sets; A multi-channel feature map construction module, used to calculate the number of features of each integrated data in the integrated data set, and construct a multi-channel feature map of the integrated data according to the number of features; A data classification module, used for calculating the target data feature of each of the integrated data according to the multi-channel feature map, and classifying the integrated data set according to the target data feature to obtain a classified data set; A related data index creation module, used to determine the related data of each integrated data according to the classified data set, and create a related data index according to the related data; The data query module is used to parse the target data of the data query request according to the associated data index when receiving the data query request.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the data query method for data islands as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the data query method for data islands as described in any one of claims 1 to 7 is implemented.

Images