CN117891961B - Data cascade sharing method and system based on map product aggregation - Google Patents
Data cascade sharing method and system based on map product aggregation Download PDFInfo
- Publication number
- CN117891961B CN117891961B CN202410288409.1A CN202410288409A CN117891961B CN 117891961 B CN117891961 B CN 117891961B CN 202410288409 A CN202410288409 A CN 202410288409A CN 117891961 B CN117891961 B CN 117891961B
- Authority
- CN
- China
- Prior art keywords
- map
- slice
- information
- culture
- history
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000002776 aggregation Effects 0.000 title claims abstract description 35
- 238000004220 aggregation Methods 0.000 title claims abstract description 35
- 238000000034 method Methods 0.000 title claims abstract description 33
- 230000004927 fusion Effects 0.000 claims abstract description 36
- 238000003860 storage Methods 0.000 claims abstract description 14
- 238000000605 extraction Methods 0.000 claims description 13
- 230000009467 reduction Effects 0.000 claims description 13
- 238000004458 analytical method Methods 0.000 claims description 12
- 238000010276 construction Methods 0.000 claims description 12
- 238000012216 screening Methods 0.000 claims description 12
- 238000012876 topography Methods 0.000 claims description 10
- 238000005065 mining Methods 0.000 claims description 4
- 230000000694 effects Effects 0.000 abstract description 8
- 238000005516 engineering process Methods 0.000 abstract description 4
- 238000013527 convolutional neural network Methods 0.000 description 7
- 238000012549 training Methods 0.000 description 6
- 238000013507 mapping Methods 0.000 description 3
- 238000007781 pre-processing Methods 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000004140 cleaning Methods 0.000 description 2
- 238000002372 labelling Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000013526 transfer learning Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Landscapes
- Processing Or Creating Images (AREA)
Abstract
The application discloses a data cascade sharing method and a system based on map product aggregation, belonging to the field of geographic information, wherein the method comprises the following steps: carrying out regional boundary slicing on a standard map, carrying out three-dimensional fusion on each map slice, generating an aggregate three-dimensional map slice, and configuring map culture labels and map history labels; the method comprises the steps of taking an aggregate three-dimensional map slice as a storage center entity, taking each map slice and a label as storage star-shaped entities, and constructing an index knowledge graph; and replacing the initial map slice with a storage center entity to obtain a map product sharing model, and building a map product sharing platform. The application solves the technical problems of low execution efficiency and high error rate caused by the retrieval and the calling of massive map slice data in the existing data sharing application of map product aggregation, and achieves the technical effects of realizing multi-level aggregation sharing of map products, improving the execution efficiency and reducing the error through the map intelligent slice and knowledge graph technology.
Description
Technical Field
The invention relates to the field of geographic information, in particular to a data cascade sharing method and system based on map product aggregation.
Background
Digital map products are widely used in various fields, and in order to meet the application demands of users in different industries, the map data are often required to be subjected to multi-level aggregation sharing at present. For example, in a smart city management platform, a plurality of levels of a plurality of types of map products such as a standard map, a traffic map, a pipeline map, and the like are required to be aggregated. However, in the process of map data aggregation sharing application, the execution efficiency is low due to the need of processing the retrieval and the call of massive map slice files; meanwhile, in the prior art, if an automatic map slice is performed according to coordinates, errors are likely to occur when the slice with a large calculation amount is faced, and the error rate is high.
Disclosure of Invention
The application provides a data cascade sharing method and a system based on map product aggregation, which aim to solve the technical problems of low execution efficiency and high error rate caused by the retrieval and the calling of massive map slice data in the existing map product aggregation data sharing application.
In view of the above problems, the present application provides a data cascade sharing method and system based on map product aggregation.
In a first aspect of the disclosure, a method for sharing data cascade based on aggregation of map products is provided, the method comprising: carrying out regional boundary slicing on a standard map through an intelligent slice assembly to obtain a first map slice, wherein the first map slice belongs to a first map slice tree, any node of the first map slice tree uniquely corresponds to one map slice, the map slice corresponding to an upper node comprises the map slice corresponding to a lower node, and regional name labels of the map slices corresponding to the lower node are stored on the edges from the upper node to the lower node; traversing a topography map, a vegetation map and a building map to carry out slicing according to the first map slice, and obtaining a second map slice, a third map slice and a fourth map slice; three-dimensional fusion is carried out on the first map slice, the second map slice, the third map slice and the fourth map slice through the map fusion component, so that a first aggregate three-dimensional map slice is generated; configuring a map culture label and a map history label of the first aggregate three-dimensional map slice; constructing a first index knowledge map by taking a first aggregate three-dimensional map slice as a first storage center entity and taking a first map slice, a second map slice, a third map slice, a fourth map slice, a map culture label and a map history label as storage star-shaped entities; traversing the first map slice tree for repeated analysis to obtain a second index knowledge graph until an N index knowledge graph is obtained, wherein N is an integer, and N is more than or equal to 1; and replacing the initial map slices of the N nodes of the first map slice tree with a first storage center entity of the first index knowledge graph and a second storage center entity of the second index knowledge graph until an N storage center entity of the N index knowledge graph to obtain a map product sharing model, and building a map product sharing platform based on the map product sharing model.
In another aspect of the disclosure, a data cascade sharing system based on map product aggregation is provided, the system comprising: the regional boundary slicing module is used for carrying out regional boundary slicing on the standard map through the intelligent slicing component to obtain a first map slice, wherein the first map slice belongs to a first map slice tree, any node of the first map slice tree uniquely corresponds to one map slice, the map slice corresponding to an upper node comprises the map slice corresponding to a lower node, and regional name labels of the map slices corresponding to the lower node are stored on the edges from the upper node to the lower node; the traversing map section module is used for traversing the topography map, the vegetation map and the building map to carry out section according to the first map section to obtain a second map section, a third map section and a fourth map section; the slice three-dimensional fusion module is used for carrying out three-dimensional fusion on the first map slice, the second map slice, the third map slice and the fourth map slice through the map fusion component to generate a first aggregate three-dimensional map slice; the label configuration module is used for configuring map culture labels and map history labels of the first aggregate three-dimensional map slices; the knowledge map construction module is used for constructing a first index knowledge map by taking a first aggregate three-dimensional map slice as a first storage center entity and taking a first map slice, a second map slice, a third map slice, a fourth map slice, a map culture label and a map history label as storage star-shaped entities; the slice tree traversing analysis module is used for traversing the first map slice tree to carry out repeated analysis to obtain a second index knowledge graph until an N index knowledge graph is obtained, wherein N is an integer, and N is more than or equal to 1; the sharing platform building module is used for replacing the initial map slices of the N nodes of the first map slice tree with a first storage center entity of the first index knowledge graph and a second storage center entity of the second index knowledge graph until an N storage center entity of the N index knowledge graph to obtain a map product sharing model, and building the map product sharing platform based on the map product sharing model.
One or more technical schemes provided by the application have at least the following technical effects or advantages:
The intelligent slicing component is used for carrying out area boundary intelligent slicing on the standard map, so that a first map slicing tree is obtained, and slicing of the standard map is efficiently and accurately completed; traversing other types of maps to carry out corresponding area slicing according to the first map slicing tree, so as to obtain slices corresponding to other types of maps corresponding to the standard map; three-dimensional scene fusion is carried out on all the slices by adopting a map fusion component, a first aggregate three-dimensional map slice is generated, and three-dimensional aggregation of multi-source map data is realized; configuring map culture labels and map history labels for the first aggregate three-dimensional map slices so as to enrich semantic information of the slices; taking a first aggregate three-dimensional map slice as a first storage center entity, constructing a first index knowledge graph, and realizing sharing of map aggregate data; traversing a first map slice tree, constructing a plurality of index knowledge maps, and realizing multi-level aggregation sharing of map slices; the method comprises the steps of replacing an initial map slice of a map slice tree with a storage center entity, obtaining a map product sharing model, building a map product sharing platform, and completing the technical scheme of data cascade sharing, so that the technical problems of low execution efficiency and high error rate caused by retrieval and calling of massive map slice data in the existing map product aggregation data sharing application are solved, and the technical effects of realizing multi-level aggregation sharing of map products, improving execution efficiency and reducing errors through a map intelligent slice and knowledge graph technology are achieved.
The foregoing description is only an overview of the present application, and is intended to be implemented in accordance with the teachings of the present application in order that the same may be more clearly understood and to make the same and other objects, features and advantages of the present application more readily apparent.
Drawings
FIG. 1 is a schematic flow chart of a data cascade sharing method based on map product aggregation according to an embodiment of the present application;
fig. 2 is a schematic flow chart of obtaining a map history tag in a data cascade sharing method based on map product aggregation according to an embodiment of the present application;
fig. 3 is a schematic structural diagram of a data cascade sharing system based on map product aggregation according to an embodiment of the present application.
Reference numerals illustrate: the system comprises a region boundary slicing module 11, a traversing map section module 12, a slice three-dimensional fusion module 13, a label configuration module 14, a knowledge map construction module 15, a slice tree traversing analysis module 16 and a sharing platform construction module 17.
Detailed Description
The technical scheme provided by the application has the following overall thought:
The embodiment of the application provides a data cascade sharing method and a system based on map product aggregation, which are characterized in that firstly, a slicing tree of multi-level map data is constructed, then intelligent slicing, three-dimensional scene fusion and knowledge graph technology means are utilized to realize intelligent slicing of the map data, three-dimensional aggregation and cascade sharing of multi-source data, and then a map product sharing model is obtained by replacing nodes of the slicing tree, and a map product sharing platform is constructed, so that the execution efficiency is improved and the error is reduced.
Specifically, firstly, the standard map is subjected to regional boundary slicing through the intelligent slicing component to obtain a first map slice, and the slicing scheme adopted in the method is different from the traditional fixed-size slicing mode, but is based on the intelligent slicing of the convolutional neural network, so that the efficiency is improved and the error is reduced. On the basis, corresponding slices of other types of maps are realized, three-dimensional fusion is utilized, multidimensional aggregation of map slices is realized, a first aggregate three-dimensional map slice is obtained, and cultural tags and historical tags are configured for the first aggregate three-dimensional map slice so as to enrich semantic information of the map slices. And then, constructing an index knowledge graph aiming at each aggregated three-dimensional map slice in the first map slice tree, associating and sharing data of the multi-level slices, and replacing slice tree nodes to obtain a map product sharing model, so that a map product sharing platform is constructed, and the execution efficiency of map aggregated products is improved.
Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described in detail with reference to the accompanying drawings.
Embodiment one:
As shown in fig. 1, an embodiment of the present application provides a data cascade sharing method based on map product aggregation, which includes:
Carrying out regional boundary slicing on a standard map through an intelligent slicing component to obtain a first map slice, wherein the first map slice belongs to a first map slice tree, any node of the first map slice tree uniquely corresponds to one map slice, the map slice corresponding to an upper node comprises the map slice corresponding to a lower node, and regional name labels of the map slices corresponding to the lower node are stored on the edges from the upper node to the lower node;
Further, the embodiment of the application further comprises:
the intelligent slice assembly comprises a boundary positioning channel, a characteristic extraction channel and a characteristic restoration channel;
carrying out boundary identification on the standard map through the boundary positioning channel to obtain a boundary identification track set, wherein the boundary identification track set comprises relation information and a regional name label set;
traversing the boundary identification track set through the feature extraction channel to extract boundary coordinates and extracting N slice feature sets;
traversing the N slice feature sets through the feature reduction channel to perform tile reduction to obtain N map slices;
and constructing the first map slice tree according to the N map slices, the containing relation information and the regional name label set, wherein the first map slice belongs to the first map slice tree.
In the embodiment of the application, the intelligent slicing component is used for carrying out intelligent analysis and slicing on the map image and comprises a boundary positioning channel, a characteristic extraction channel and a characteristic restoration channel, and boundary identification, coordinate extraction and contour restoration of the map are realized through a convolutional neural network. The boundary positioning channel realizes automatic positioning and identification of the regional boundary of the standard map through a convolutional neural network, and inputs the regional boundary into the standard map and outputs the standard map as the identified boundary information; the input of the feature extraction channel is a standard map which is output by a boundary positioning channel and contains boundary information, the channel traverses a boundary identifier, and a boundary coordinate point set is extracted to generate a slice feature set which represents the feature of a map slice region; the input of the feature restoration channel is a slice feature set output by the feature extraction channel, the channel restores slice features by using a convolutional neural network, and map slices corresponding to the shape and the outline of the region are output.
The first map slice belongs to a first map slice tree. The first map slice tree is a hierarchical tree data structure for storing map slices and their logical relationships. Each node on the first map slice tree uniquely corresponds to a map slice. The map slice corresponding to the upper node contains the map slice corresponding to the lower node. And the edges from the upper node to the lower node store the regional name labels of the map slices corresponding to the lower node, and the regional name labels are used for identifying the regional range.
Specifically, in order to construct a first map slice tree, firstly, a standard map is input into a boundary positioning channel, the boundary positioning channel automatically identifies regional boundaries in the standard map based on a convolutional neural network, and outputs a spatial coordinate form of the boundaries, which is called a boundary identification track, and then all the extracted boundary identification tracks are collected to form a boundary identification track set which contains contour coordinate information of all regional boundaries in the standard map. And when the boundary identification track is extracted, the boundary positioning channel identifies the map region types represented by different closed regions, such as provinces, cities and the like, and outputs corresponding region names to form a region name tag set. And judging the inclusion relation between different areas according to the inclusion and crossing relation between the output different area boundaries, thereby determining the inclusion relation information between the areas. And secondly, the feature extraction channel sequentially takes out each boundary identification track in the boundary identification track set in a traversing way, analyzes the coordinate point set of the boundary identification track, extracts the space coordinate features such as the area of the region, the perimeter of the outline, the length of the boundary line segment, the coordinates of the center point and the like, and forms a slice feature set representing the closed region feature of the boundary identification track, so that N slice feature sets are obtained and correspond to the boundary identification tracks contained in the boundary identification track set one by one.
And then, traversing the obtained N slice feature sets by the feature reduction channel, sequentially taking out the slice feature sets, carrying out combined reduction on each tile in the slice according to the information such as the area, the contour line segment and the like contained in the slice feature sets, and generating corresponding and matched image area slices, namely map slices, so as to obtain N map slices, and realizing pixel-level segmentation of a standard map in a one-to-one correspondence manner with the N slice feature sets. Then, the containing relation information is abstracted into a hierarchical tree structure of the map area, and the upper and lower levels and the containing relation between different areas are determined according to the containing relation described therein. And simultaneously, the N map slices are in one-to-one correspondence with the regional name label sets, and the different map slices correspond to a certain level region in the tree structure. Then, the map slice is regarded as a tree node, and the parent-child cascade relation among the slice nodes is established according to the inclusion relation among the areas. And repeatedly connecting all map slices according to the parent-child cascade relation, and expanding the map slices into a first map slice tree. The first map slice is used as a leaf node of the first map slice tree, sequentially corresponds to different levels and belongs to a part of the first map slice tree.
Through constructing the map slice tree structure with hierarchical relation, when the area searched by the user belongs to a certain province-certain city-certain area, the searching starting point can be directly determined from the province, then the next layer of nodes are determined from the city of the province, and then the area is rapidly determined, so that rapid positioning and local searching are realized. Compared with tiles equally divided according to preset sizes in the prior art, the map slicing tree has no area tag, the storage structure is pyramid-shaped, and the retrieval efficiency can be greatly improved.
Further, the boundary locating channel constructing step includes: collecting a standard map data set and a regional boundary identification data set, taking the regional boundary identification data set as supervision data, taking the standard map data set as input data, and configuring the boundary positioning channel;
In a preferred embodiment, to configure a boundary locating channel, firstly, a plurality of digitized standard map files are acquired from a map publishing company or a mapping department, a standard map data set is acquired, region boundary locating and marking are manually carried out on the acquired standard map, region boundary lines of different levels are extracted, and files generated by marking boundaries are stored to acquire a region boundary identification data set. And then, preprocessing such as data cleaning, sampling, formatting conversion and the like is carried out on the collected standard map data set and the region boundary identification data set, and image input and labeling target output required by training are constructed. And then, selecting a convolutional neural network model framework, constructing a model of a convolutional coding and decoding structure, enabling an input layer to correspond to standard map image data, and enabling an output layer to output a boundary identification image. Then, the preprocessed standard map data is used as an input sample, the corresponding area boundary identification data is used as a supervision label, parameters of a network model are trained, a classification cross entropy loss function is used as an optimization target, and a gradient descent method is used for updating network parameters. And then, testing the boundary positioning effect of the model, continuously optimizing the network structure and parameters according to the effect, and finally setting the model framework and the parameters thereof which are obtained through training as boundary positioning channels, so as to complete the configuration of the boundary positioning channels and realize the regional boundary positioning of the input standard map.
Further, the feature reduction channel construction step includes: and collecting a region coordinate feature set and a region contour identification data set, taking the region contour identification data set as supervision data, taking the region coordinate feature set as input data, and configuring the feature restoration channel.
In a preferred embodiment, to configure the feature restoration channel, first, a plurality of sample map slices are collected by a map publishing or mapping department, region boundary coordinates in the plurality of sample map slices are loaded and analyzed, the area, perimeter, boundary line segment length, center point and the like of each region are calculated as region coordinate features, and the plurality of sample map slices are repeatedly processed to form a region coordinate feature set. And secondly, manually marking contours of a plurality of acquired sample map slices to form a region contour identification data set. And then, preprocessing such as data cleaning, pairing, formatting conversion and the like is carried out on the acquired region coordinate feature set and region contour identification data set, so as to construct input and output required by training. And then selecting and generating an countermeasure network frame, constructing a model structure of a generator and a discriminator, inputting the region coordinate characteristics corresponding to the layer, and generating a corresponding region contour image by the output layer. And then, taking the preprocessed region coordinate characteristics as input samples, taking corresponding region contour identification data as a supervision tag, training parameters of a network model, taking the loss resistance and the reconstruction loss as optimization targets, and updating the parameters by using a gradient method. And then, the test model generates the effect of the region contour mark from the region coordinate characteristics, and optimizes the network structure and the super parameters according to the effect, so that a characteristic reduction channel is obtained, and the configuration of the characteristic reduction channel is completed.
Traversing a topography map, a vegetation map and a building map to carry out slicing according to the first map slice, and obtaining a second map slice, a third map slice and a fourth map slice;
In the embodiment of the application, firstly, a topography map, a vegetation map and a building map are loaded, and image preprocessing including coordinate transformation, cutting and the like is carried out, so that the topography map, the vegetation map and the building map are aligned with a standard map. After a first map slice of a standard map is acquired, the regional range information of the first map slice is analyzed, the spatial information such as coordinate boundaries and areas of the regional range information is extracted, the range of the first map slice is projected onto a topographic map, a corresponding region is intercepted, and a topographic map slice is generated and used as a second map slice. And similarly, intercepting the range corresponding to the first map slice on the vegetation map and the building map in turn, and generating a corresponding area slice to obtain a third map slice and a fourth map slice.
Corresponding map slices are synchronously acquired on a topography map, a vegetation map and a building map according to the first map slice, so that automatic generation of multi-type map collaborative slices is realized, and a foundation is provided for fusion of multi-source heterogeneous map data.
Three-dimensional fusion is carried out on the first map slice, the second map slice, the third map slice and the fourth map slice through a map fusion component, so that a first aggregate three-dimensional map slice is generated;
further, before this step, the method further includes:
collecting a plurality of standard maps, a plurality of topographic maps, a plurality of vegetation maps, a plurality of building maps and a plurality of three-dimensional identification models of a plurality of areas, wherein any one area of the plurality of areas has a reference constraint size;
Taking the plurality of three-dimensional identification models as supervision, taking the plurality of standard maps, the plurality of topographic maps, the plurality of vegetation maps and the plurality of building maps as input, and configuring the map fusion assembly.
In one possible embodiment, firstly, a plurality of areas available for training are determined, and standard maps, topography maps, vegetation maps and building maps of the plurality of areas are obtained through a map publishing company or a mapping department, so that a plurality of standard maps, a plurality of topography maps, a plurality of vegetation maps and a plurality of building maps are obtained. Meanwhile, various map sizes are controlled to be consistent by the reference constraint size, and a foundation is provided for configuring the map fusion component. Then, high-precision three-dimensional scenes of a plurality of areas are collected and constructed to serve as ground truth values through unmanned aerial vehicle aerial survey and other means, and a plurality of three-dimensional identification models are obtained. Then, a map coding network represented by a convolutional neural network and a three-dimensional convolutional decoding network are constructed, the map coding network inputs various maps, and the three-dimensional convolutional decoding network outputs a three-dimensional scene. And then, training the three-dimensional convolution decoding network parameters by using a data set such as image recognition and the like through transfer learning to enable the coding network to learn map feature extraction, using a plurality of standard maps, a plurality of topographic maps, a plurality of vegetation maps and a plurality of building maps as input and using a plurality of three-dimensional identification models as supervision labels to perform three-dimensional fusion migration. And then, connecting the trained map coding network and the three-dimensional convolution decoding network into a map fusion assembly, and realizing the configuration of the map fusion assembly.
After a first map slice, a second map slice, a third map slice and a fourth map slice are obtained, the first map slice, the second map slice, the third map slice and the fourth map slice are input into a map fusion component together, the map fusion component analyzes image features of different types of map slices through a map coding network, performs feature alignment fusion, and then generates an aggregated three-dimensional scene through a three-dimensional convolution decoding network to obtain a first aggregated three-dimensional map slice.
Configuring a map culture label and a map history label of the first aggregate three-dimensional map slice;
further, as shown in fig. 2, the embodiment of the present application further includes:
obtaining a first area name tag of the first aggregate three-dimensional map slice;
searching the used regional name set according to the first regional name label to construct a first regional name set;
Traversing the first regional name set to carry out map culture and map history frequent mining to obtain an initial map culture information set and an initial map history information set;
and carrying out time sequence arrangement on the initial map culture information set to obtain the map culture label, and carrying out time sequence arrangement on the initial map history information set to obtain the map history label.
Further, the embodiment of the application further comprises:
Traversing the first regional name set to perform map culture and map history retrieval to obtain a map culture retrieval information set and a map history retrieval information set;
Traversing the map culture retrieval information set, counting a first information on-line reaching frequency feature set, traversing the map history retrieval information set, and counting a second information on-line reaching frequency feature set, wherein the first information on-line reaching frequency feature set and the second information on-line reaching frequency feature set represent detected frequency numbers of corresponding information on a network;
Screening map culture retrieval information with the touch frequency characteristic greater than or equal to a touch frequency characteristic threshold value on the first information line from the map culture retrieval information set according to the touch frequency characteristic set on the first information line, and adding the map culture retrieval information into the initial map culture information set;
And screening map history retrieval information with the frequency of touch on the second information line greater than or equal to a frequency of touch feature threshold value from the map history retrieval information set according to the frequency of touch on the second information line, and adding the map history retrieval information into the initial map history information set.
In one possible implementation, first, metadata of the first aggregate three-dimensional map slice is loaded, and attribute information such as a name, a management number, a geographic coordinate range and the like of the first aggregate three-dimensional map slice is resolved from the metadata. Secondly, words or phrases representing the region range are extracted from the attributes of the first aggregate three-dimensional map slice through a rule matching or character string similarity algorithm, and place name recognition is carried out, so that place name entries appearing therein, such as landmark building names appearing therein, are taken. And then, combining the obtained place names and the attribute information in series to form descriptive text expressing the area range represented by the first aggregate three-dimensional map slice, and defining the descriptive text as a first area name label. And then, carrying out fuzzy matching query in a historical place name database by using the first area name label as a keyword, and extracting the unique names and the ancient names related to the area represented by the first area name label from the query result to form a first area name set.
And traversing the first regional name set, taking out each name, combining the name and the map culture as query words, carrying out network retrieval based on a search engine API, and storing the returned map culture related pages as retrieval results to form a map culture retrieval information set. And similarly, combining the name with the map history category to perform combined query, and acquiring related pages to form a map history retrieval information set. And then, analyzing the map culture retrieval information set one by one, extracting the characteristics of keywords, source websites, page URLs and the like of each information, taking each page URL as a seed, acquiring the frequency quantity characteristics of the page such as the network accumulated access quantity, the number of access users and the like by utilizing the statistical API of the network search engine, and associating the frequency quantity characteristics to the corresponding map culture information to form a first information online touch frequency characteristic set in an aggregation mode. And similarly, traversing each result of the map history retrieval information set, acquiring the frequency quantity characteristics of page access, and constructing a second information on-line reaching frequency characteristic set. And then preselecting a network touch frequency characteristic threshold value of map culture information, for example, the total page access amount needs to reach 1000 times, traversing the touch frequency characteristic set on the first information line, reading the statistical access amount of each piece of map culture retrieval information, comparing the access amount with the touch frequency characteristic threshold value, selecting the map culture retrieval information with the access amount larger than or equal to the touch frequency characteristic threshold value, and adding the map culture retrieval information to the initial map culture information set for storage. And traversing the touch frequency feature set on the second information line, reading the statistical access quantity of each piece of map history retrieval information, comparing the access quantity with the touch frequency feature threshold, selecting the map history retrieval information with the access quantity being greater than or equal to the touch frequency feature threshold, and adding the map history retrieval information to the initial map history information set for storage.
Then, the initial map culture information set is loaded, and timing key information of each map culture information item, such as a time interval of culture or remains, is analyzed. And then, extracting the starting year and the ending year of each map culture information item by using a time information extraction technology, sequencing and arranging the initial map culture information set according to the size of the years, and combining the map culture information sets into a map culture label according to sequential and consecutive arrangement sequences to reflect the time sequence of the regional culture history. And similarly, the initial map history information sets are arranged in time sequence to form map history labels, and the time sequence of the history event evolution is reflected. And then, respectively associating and storing the map culture label and the map history label with corresponding map slices, and intuitively reflecting the time dimension evolution of the regional culture and the history.
By the method, map culture labels and map history labels are automatically configured for all areas, and the defects that the efficiency is low and information is incomplete due to manual labeling of all areas are avoided, so that the map culture labels and the map history labels are configured for all the areas comprehensively and efficiently.
Further, before counting the frequency feature set of the touch on the first information line and the frequency feature set of the touch on the second information line, the embodiment of the application further includes:
The map culture retrieval information set is provided with a first information source label set, and the map history retrieval information set is provided with a second information source label set;
Invoking a list of trusted sources;
Screening map culture retrieval information belonging to the trusted source list in the first information source tag set from the map culture retrieval information set, and adding the map culture retrieval information into the initial map culture information set;
And screening the map history retrieval information belonging to the trusted source list in the second information source tag set from the map history retrieval information set, and adding the map history retrieval information into the initial map history information set.
In a preferred embodiment, before counting the first information line reaching frequency feature set and the second information line reaching frequency feature set, first, traversing the map culture retrieval information set, identifying the release source of each map culture retrieval information, such as a news website, forum, and the like, and generating a first information source label set describing the source of each map culture retrieval information. In the same way, the map history retrieval information set is traversed, and the map history retrieval information sources are extracted to construct a second information source tag set. Then, a trusted source list is extracted, wherein the trusted source list comprises authoritative information release sources such as news media, academic theory libraries and the like which are evaluated by experts.
And then traversing the map culture retrieval information set, reading a first information source label corresponding to each map culture retrieval information, carrying out matching check on the first information source label and the trusted source list, and if the first information source label belongs to the trusted source list, judging that the information quality is higher, and adding the map culture retrieval information into the initial map culture information set for storage. And similarly, traversing the map history retrieval information set, reading a second information source label corresponding to each map history retrieval information, performing matching check on the second information source label and the trusted source list, and if the second information source label belongs to the trusted source list, judging that the information quality is higher, adding the map history retrieval information to the initial map history information set for storage.
By matching with the trusted source list, the quality judgment of the map culture information and the map history information is automatically realized, and the construction efficiency and the credibility of the initial map culture information set and the initial map history information set are improved.
Constructing a first index knowledge graph by taking the first aggregate three-dimensional map slice as a first storage center entity and taking the first map slice, the second map slice, the third map slice, the fourth map slice, the map culture label and the map history label as storage star-shaped entities;
in the embodiment of the application, firstly, a model framework of a knowledge graph is created, the framework takes an aggregate three-dimensional map slice as a storage center entity, map slices of a standard map, a topographic map, a vegetation map and a building map are connected around the storage center entity, and simultaneously, a map culture label and a map history label corresponding to the aggregate three-dimensional map slice are connected to form a star-mounted knowledge graph structure. After the first aggregate three-dimensional map slice is obtained, the first aggregate three-dimensional map slice is taken as a storage center entity, and the first map slice, the second map slice, the third map slice and the fourth map slice corresponding to the first aggregate three-dimensional map slice, and the map culture label and the map history label corresponding to the first aggregate three-dimensional map slice are stored in association with the first aggregate three-dimensional map slice to form a storage star-shaped entity, so that the construction of the first index knowledge graph is completed.
Traversing the first map slicing tree for repeated analysis to obtain a second index knowledge graph until an N index knowledge graph is obtained, wherein N is an integer, and N is more than or equal to 1;
In the embodiment of the application, after a first index knowledge graph is built by a first map slice in a first map slice tree, other map slices in the first map slice tree are continuously traversed, and initial map slices corresponding to each node in the first map slice tree are sequentially taken out. And then, repeatedly performing map aggregation on each initial map slice to generate an aggregate three-dimensional map slice, configuring map culture labels and map history labels and constructing an index knowledge graph, recursively traversing the process to an N-th map slice of a first map slice tree all the time, and repeating the steps to obtain a series of index knowledge graphs corresponding to each node of the first map slice tree, wherein the series of index knowledge graphs are expressed as a second index knowledge graph to an N-th index knowledge graph. And N is an integer, and the value of N is greater than or equal to 1, and represents the node number of the first map slice tree.
By taking the first map slice tree as the basic framework for supporting and then adopting an automatic iteration mode, the index knowledge graph of the map slice corresponding to each node is efficiently generated, the multidimensional map knowledge expression is realized, and the support is provided for data cascade sharing.
And replacing the initial map slices of the N nodes of the first map slice tree with a first storage center entity of the first index knowledge graph and a second storage center entity of the second index knowledge graph until an N storage center entity of the N index knowledge graph to obtain a map product sharing model, and building a map product sharing platform based on the map product sharing model.
In the embodiment of the application, first, a first map slice tree and an obtained first index knowledge graph are loaded until an Nth index knowledge graph. And then, replacing the original initial map slices of N nodes in the first map slice tree with storage center entities corresponding to the index knowledge maps. For example, a first map slice in a first node in a first map slice tree is replaced with a first storage center entity of a first index knowledge-graph. And after the initial map slices of all the nodes in the first map slice tree are replaced by the corresponding storage center entities of the index knowledge graph, obtaining a map tree with rich knowledge as a map product sharing model. And developing application interface services such as map retrieval, map information display and the like on the basis of the map product sharing model, and adopting a distributed network technology to realize deployment and calling of the services so as to build a map product sharing platform.
Through the organic fusion of the map slicing tree and the knowledge graph, the knowledge adjacency and enhancement of map data are realized, and the intelligent cascade sharing and utilization of map information are supported, so that the execution efficiency of the aggregate map is improved, and the error of the aggregate map data is reduced.
In summary, the data cascade sharing method based on map product aggregation provided by the embodiment of the application has the following technical effects:
And the intelligent slicing component performs regional boundary slicing on the standard map to obtain a first map slice, so that intelligent slicing of the map is efficiently and accurately completed. And traversing the topographic map, the vegetation map and the building map to slice according to the first map slice to obtain a second map slice, a third map slice and a fourth map slice, and obtaining slices corresponding to other types of maps. And carrying out three-dimensional fusion on the first map slice, the second map slice, the third map slice and the fourth map slice through a map fusion component to generate a first aggregate three-dimensional map slice, so as to realize three-dimensional scene aggregation of map multi-source data. And configuring a map culture label and a map history label of the first aggregate three-dimensional map slice, and enriching semantic information of the map slice. And building a first index knowledge graph by taking the first aggregate three-dimensional map slice as a first storage center entity and taking the first map slice, the second map slice, the third map slice, the fourth map slice, the map culture label and the map history label as storage star-shaped entities, so as to realize the association and sharing of the data after the map aggregation. And traversing the first map slice tree to perform repeated analysis, and obtaining a second index knowledge graph until an N index knowledge graph is obtained, wherein N is an integer, and N is more than or equal to 1, so that cascade sharing of data of the multi-level map slices is realized. And replacing the initial map slices of the N nodes of the first map slice tree with a first storage center entity of the first index knowledge graph and a second storage center entity of the second index knowledge graph until an N storage center entity of the N index knowledge graph to obtain a map product sharing model, building a map product sharing platform based on the map product sharing model, and completing data cascade sharing of map product aggregation.
Embodiment two:
based on the same inventive concept as the data cascade sharing method based on map product aggregation in the foregoing embodiment, as shown in fig. 3, an embodiment of the present application provides a data cascade sharing system based on map product aggregation, including:
The regional boundary slicing module 11 is configured to perform regional boundary slicing on a standard map through the intelligent slicing component to obtain a first map slice, where the first map slice belongs to a first map slice tree, any node of the first map slice tree uniquely corresponds to one map slice, the map slice corresponding to an upper node includes the map slice corresponding to a lower node, and regional name labels of the map slices corresponding to the lower node are stored on edges from the upper node to the lower node;
A traversing map slicing module 12, configured to slice a terrain map, a vegetation map, and a building map according to the first map slice, and obtain a second map slice, a third map slice, and a fourth map slice;
the slice three-dimensional fusion module 13 is configured to generate a first aggregated three-dimensional map slice by performing three-dimensional fusion on the first map slice, the second map slice, the third map slice, and the fourth map slice through a map fusion component;
a label configuration module 14 for configuring map culture labels and map history labels of the first aggregate three-dimensional map slice;
The knowledge graph construction module 15 is configured to construct a first index knowledge graph by using the first aggregated three-dimensional map slice as a first storage center entity, and using the first map slice, the second map slice, the third map slice, the fourth map slice, the map culture label, and the map history label as storage star-shaped entities;
The slice tree traversal analysis module 16 is configured to traverse the first map slice tree for repeated analysis, obtain a second index knowledge graph until an nth index knowledge graph, where N is an integer, and N is greater than or equal to 1;
The sharing platform building module 17 is configured to replace an initial map slice of N nodes of the first map slice tree with a first storage center entity of the first index knowledge graph and an nth storage center entity of the second index knowledge graph until the nth storage center entity of the nth index knowledge graph, obtain a map product sharing model, and build a map product sharing platform based on the map product sharing model.
Further, the region boundary slicing module 11 includes the following steps:
the intelligent slice assembly comprises a boundary positioning channel, a characteristic extraction channel and a characteristic restoration channel;
carrying out boundary identification on the standard map through the boundary positioning channel to obtain a boundary identification track set, wherein the boundary identification track set comprises relation information and a regional name label set;
traversing the boundary identification track set through the feature extraction channel to extract boundary coordinates and extracting N slice feature sets;
traversing the N slice feature sets through the feature reduction channel to perform tile reduction to obtain N map slices;
and constructing the first map slice tree according to the N map slices, the containing relation information and the regional name label set, wherein the first map slice belongs to the first map slice tree.
Further, the region boundary slicing module 11 further includes the following steps:
The boundary locating channel construction step comprises the following steps: collecting a standard map data set and a regional boundary identification data set, taking the regional boundary identification data set as supervision data, taking the standard map data set as input data, and configuring the boundary positioning channel;
The feature reduction channel construction step includes: and collecting a region coordinate feature set and a region contour identification data set, taking the region contour identification data set as supervision data, taking the region coordinate feature set as input data, and configuring the feature restoration channel.
Further, the embodiment of the application also comprises a map fusion component configuration module, which comprises the following execution steps:
collecting a plurality of standard maps, a plurality of topographic maps, a plurality of vegetation maps, a plurality of building maps and a plurality of three-dimensional identification models of a plurality of areas, wherein any one area of the plurality of areas has a reference constraint size;
Taking the plurality of three-dimensional identification models as supervision, taking the plurality of standard maps, the plurality of topographic maps, the plurality of vegetation maps and the plurality of building maps as input, and configuring the map fusion assembly.
Further, the tag configuration module 14 includes the following execution steps:
obtaining a first area name tag of the first aggregate three-dimensional map slice;
searching the used regional name set according to the first regional name label to construct a first regional name set;
Traversing the first regional name set to carry out map culture and map history frequent mining to obtain an initial map culture information set and an initial map history information set;
and carrying out time sequence arrangement on the initial map culture information set to obtain the map culture label, and carrying out time sequence arrangement on the initial map history information set to obtain the map history label.
Further, the tag configuration module 14 further includes the following steps:
Traversing the first regional name set to perform map culture and map history retrieval to obtain a map culture retrieval information set and a map history retrieval information set;
Traversing the map culture retrieval information set, counting a first information on-line reaching frequency feature set, traversing the map history retrieval information set, and counting a second information on-line reaching frequency feature set, wherein the first information on-line reaching frequency feature set and the second information on-line reaching frequency feature set represent detected frequency numbers of corresponding information on a network;
Screening map culture retrieval information with the touch frequency characteristic greater than or equal to a touch frequency characteristic threshold value on the first information line from the map culture retrieval information set according to the touch frequency characteristic set on the first information line, and adding the map culture retrieval information into the initial map culture information set;
And screening map history retrieval information with the frequency of touch on the second information line greater than or equal to a frequency of touch feature threshold value from the map history retrieval information set according to the frequency of touch on the second information line, and adding the map history retrieval information into the initial map history information set.
Further, the embodiment of the application further comprises a retrieval information set acquisition module, which comprises the following execution steps:
The map culture retrieval information set is provided with a first information source label set, and the map history retrieval information set is provided with a second information source label set;
Invoking a list of trusted sources;
Screening map culture retrieval information belonging to the trusted source list in the first information source tag set from the map culture retrieval information set, and adding the map culture retrieval information into the initial map culture information set;
And screening the map history retrieval information belonging to the trusted source list in the second information source tag set from the map history retrieval information set, and adding the map history retrieval information into the initial map history information set.
Any of the steps of the methods described above may be stored as computer instructions or programs in a non-limiting computer memory and may be called by a non-limiting computer processor to identify any method for implementing an embodiment of the present application, without unnecessary limitations.
Further, the first or second element may not only represent a sequential relationship, but may also represent a particular concept, and/or may be selected individually or in whole among a plurality of elements. It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the scope of the application. Thus, the present application is intended to include such modifications and alterations insofar as they come within the scope of the application or the equivalents thereof.
Claims (8)
1. The data cascade sharing method based on map product aggregation is characterized by comprising the following steps:
Carrying out regional boundary slicing on a standard map through an intelligent slicing component to obtain a first map slice, wherein the first map slice belongs to a first map slice tree, any node of the first map slice tree uniquely corresponds to one map slice, the map slice corresponding to an upper node comprises the map slice corresponding to a lower node, and regional name labels of the map slices corresponding to the lower node are stored on the edges from the upper node to the lower node;
Traversing a topography map, a vegetation map and a building map to carry out slicing according to the first map slice, and obtaining a second map slice, a third map slice and a fourth map slice;
three-dimensional fusion is carried out on the first map slice, the second map slice, the third map slice and the fourth map slice through a map fusion component, so that a first aggregate three-dimensional map slice is generated;
configuring a map culture label and a map history label of the first aggregate three-dimensional map slice;
Constructing a first index knowledge graph by taking the first aggregate three-dimensional map slice as a first storage center entity and taking the first map slice, the second map slice, the third map slice, the fourth map slice, the map culture label and the map history label as storage star-shaped entities;
Traversing the first map slicing tree for repeated analysis to obtain a second index knowledge graph until an N index knowledge graph is obtained, wherein N is an integer, and N is more than or equal to 1;
And replacing the initial map slices of the N nodes of the first map slice tree with a first storage center entity of the first index knowledge graph and a second storage center entity of the second index knowledge graph until an N storage center entity of the N index knowledge graph to obtain a map product sharing model, and building a map product sharing platform based on the map product sharing model.
2. The method of claim 1, wherein the performing, by the intelligent slicing component, the region boundary slice on the standard map to obtain the first map slice comprises:
the intelligent slice assembly comprises a boundary positioning channel, a characteristic extraction channel and a characteristic restoration channel;
carrying out boundary identification on the standard map through the boundary positioning channel to obtain a boundary identification track set, wherein the boundary identification track set comprises relation information and a regional name label set;
traversing the boundary identification track set through the feature extraction channel to extract boundary coordinates and extracting N slice feature sets;
traversing the N slice feature sets through the feature reduction channel to perform tile reduction to obtain N map slices;
and constructing the first map slice tree according to the N map slices, the containing relation information and the regional name label set, wherein the first map slice belongs to the first map slice tree.
3. The method as claimed in claim 2, comprising:
The boundary locating channel construction step comprises the following steps: collecting a standard map data set and a regional boundary identification data set, taking the regional boundary identification data set as supervision data, taking the standard map data set as input data, and configuring the boundary positioning channel;
The feature reduction channel construction step includes: and collecting a region coordinate feature set and a region contour identification data set, taking the region contour identification data set as supervision data, taking the region coordinate feature set as input data, and configuring the feature restoration channel.
4. The method of claim 1, wherein three-dimensionally fusing the first map slice, the second map slice, the third map slice, and the fourth map slice by a map fusing component to generate a first aggregate three-dimensional map slice, the steps comprising:
collecting a plurality of standard maps, a plurality of topographic maps, a plurality of vegetation maps, a plurality of building maps and a plurality of three-dimensional identification models of a plurality of areas, wherein any one area of the plurality of areas has a reference constraint size;
Taking the plurality of three-dimensional identification models as supervision, taking the plurality of standard maps, the plurality of topographic maps, the plurality of vegetation maps and the plurality of building maps as input, and configuring the map fusion assembly.
5. The method of claim 1, wherein configuring map culture tags and map history tags for the first aggregate three-dimensional map slice comprises:
obtaining a first area name tag of the first aggregate three-dimensional map slice;
searching the used regional name set according to the first regional name label to construct a first regional name set;
Traversing the first regional name set to carry out map culture and map history frequent mining to obtain an initial map culture information set and an initial map history information set;
and carrying out time sequence arrangement on the initial map culture information set to obtain the map culture label, and carrying out time sequence arrangement on the initial map history information set to obtain the map history label.
6. The method of claim 5, wherein traversing the first set of area names for map culture and map history frequent mining obtains an initial set of map culture information and an initial set of map history information, comprising:
Traversing the first regional name set to perform map culture and map history retrieval to obtain a map culture retrieval information set and a map history retrieval information set;
Traversing the map culture retrieval information set, counting a first information on-line reaching frequency feature set, traversing the map history retrieval information set, and counting a second information on-line reaching frequency feature set, wherein the first information on-line reaching frequency feature set and the second information on-line reaching frequency feature set represent detected frequency numbers of corresponding information on a network;
Screening map culture retrieval information with the touch frequency characteristic greater than or equal to a touch frequency characteristic threshold value on the first information line from the map culture retrieval information set according to the touch frequency characteristic set on the first information line, and adding the map culture retrieval information into the initial map culture information set;
And screening map history retrieval information with the frequency of touch on the second information line greater than or equal to a frequency of touch feature threshold value from the map history retrieval information set according to the frequency of touch on the second information line, and adding the map history retrieval information into the initial map history information set.
7. The method of claim 6, wherein traversing the map culture retrieval information set, counting a frequency of touchdown feature set on a first information line, traversing the map history retrieval information set, counting a frequency of touchdown feature set on a second information line, and before;
The map culture retrieval information set is provided with a first information source label set, and the map history retrieval information set is provided with a second information source label set;
Invoking a list of trusted sources;
Screening map culture retrieval information belonging to the trusted source list in the first information source tag set from the map culture retrieval information set, and adding the map culture retrieval information into the initial map culture information set;
And screening the map history retrieval information belonging to the trusted source list in the second information source tag set from the map history retrieval information set, and adding the map history retrieval information into the initial map history information set.
8. A map product aggregation based data cascade sharing system for implementing the map product aggregation based data cascade sharing method of any one of claims 1-7, the system comprising:
The regional boundary slicing module is used for carrying out regional boundary slicing on a standard map through the intelligent slicing assembly to obtain a first map slice, wherein the first map slice belongs to a first map slice tree, any node of the first map slice tree uniquely corresponds to one map slice, the map slice corresponding to an upper node comprises the map slice corresponding to a lower node, and regional name labels of the map slices corresponding to the lower node are stored on the edges from the upper node to the lower node;
The traversing map section module is used for traversing the topography map, the vegetation map and the building map to be sectioned according to the first map section to obtain a second map section, a third map section and a fourth map section;
The slice three-dimensional fusion module is used for carrying out three-dimensional fusion on the first map slice, the second map slice, the third map slice and the fourth map slice through a map fusion component to generate a first aggregate three-dimensional map slice;
The label configuration module is used for configuring map culture labels and map history labels of the first aggregate three-dimensional map slices;
The knowledge graph construction module is used for constructing a first index knowledge graph by taking the first aggregate three-dimensional map slice as a first storage center entity and taking the first map slice, the second map slice, the third map slice, the fourth map slice, the map culture label and the map history label as storage star-shaped entities;
The slice tree traversal analysis module is used for traversing the first map slice tree to carry out repeated analysis, so as to obtain a second index knowledge graph until an N index knowledge graph is obtained, N is an integer, and N is more than or equal to 1;
The sharing platform building module is used for replacing the initial map slices of the N nodes of the first map slicing tree with the first storage center entity of the first index knowledge graph and the second storage center entity of the second index knowledge graph until the N storage center entity of the N index knowledge graph to obtain a map product sharing model, and building a map product sharing platform based on the map product sharing model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410288409.1A CN117891961B (en) | 2024-03-14 | 2024-03-14 | Data cascade sharing method and system based on map product aggregation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410288409.1A CN117891961B (en) | 2024-03-14 | 2024-03-14 | Data cascade sharing method and system based on map product aggregation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117891961A CN117891961A (en) | 2024-04-16 |
| CN117891961B true CN117891961B (en) | 2024-05-28 |
Family
ID=90643019
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410288409.1A Active CN117891961B (en) | 2024-03-14 | 2024-03-14 | Data cascade sharing method and system based on map product aggregation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117891961B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2008005761A2 (en) * | 2006-06-30 | 2008-01-10 | Healthy Interactions, Inc. | System, method, and device for providing health information |
| CN114357104A (en) * | 2022-02-25 | 2022-04-15 | 北京嘉诚瑞杰信息技术有限公司 | Non-index map slice polymerization method |
| CN116541480A (en) * | 2023-07-05 | 2023-08-04 | 中国科学院文献情报中心 | Thematic data construction method and system based on multi-label driving |
-
2024
- 2024-03-14 CN CN202410288409.1A patent/CN117891961B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2008005761A2 (en) * | 2006-06-30 | 2008-01-10 | Healthy Interactions, Inc. | System, method, and device for providing health information |
| CN114357104A (en) * | 2022-02-25 | 2022-04-15 | 北京嘉诚瑞杰信息技术有限公司 | Non-index map slice polymerization method |
| CN116541480A (en) * | 2023-07-05 | 2023-08-04 | 中国科学院文献情报中心 | Thematic data construction method and system based on multi-label driving |
Non-Patent Citations (2)
| Title |
|---|
| 基于地理空间大数据的高效索引与检索算法;赵慧慧;赵凡;陈仁海;冯志勇;;计算机研究与发展;20200215(02);全文 * |
| 基于大数据与知识图谱的知识共享服务平台;文莉莉;邬满;;电子元器件与信息技术;20200320(03);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117891961A (en) | 2024-04-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109635171B (en) | Fusion reasoning system and method for news program intelligent tags | |
| CN104317801B (en) | A kind of Data clean system and method towards big data | |
| CN110704411B (en) | Knowledge graph building method and device suitable for art field and electronic equipment | |
| CN113434623B (en) | Fusion method based on multi-source heterogeneous space planning data | |
| CN102395965B (en) | Method for searching objects in a database | |
| US11720606B1 (en) | Automated geospatial data analysis | |
| CN109710701A (en) | A kind of automated construction method for public safety field big data knowledge mapping | |
| CN112541490A (en) | Archive image information structured construction method and device based on deep learning | |
| CN114996488B (en) | Skynet big data decision-level fusion method | |
| CN106503223B (en) | online house source searching method and device combining position and keyword information | |
| CN109408578B (en) | Monitoring data fusion method for heterogeneous environment | |
| CN115934990B (en) | Remote sensing image recommendation method based on content understanding | |
| CN105045875A (en) | Personalized information retrieval method and apparatus | |
| CN104573130A (en) | Entity resolution method based on group calculation and entity resolution device based on group calculation | |
| CN104346438A (en) | Data management service system based on large data | |
| CN109165273A (en) | General Chinese address matching method facing big data environment | |
| CN113254630B (en) | Domain knowledge map recommendation method for global comprehensive observation results | |
| CN105550169A (en) | Method and device for identifying point of interest names based on character length | |
| CN111626568A (en) | Knowledge base construction method and device and knowledge search method and system | |
| CN116881430A (en) | Industrial chain identification method and device, electronic equipment and readable storage medium | |
| Belcastro et al. | Parallel extraction of Regions‐of‐Interest from social media data | |
| CN111325235B (en) | Multilingual-oriented universal place name semantic similarity calculation method and application thereof | |
| CN112800083A (en) | Government decision-oriented government affair big data analysis method and equipment | |
| CN117891961B (en) | Data cascade sharing method and system based on map product aggregation | |
| Gang | Personalized Recommendation of Tourist Attractions Based on Collaborative Filtering |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |