CN117312688B - Cross-source data retrieval method, medium and device based on space-time asset catalogue - Google Patents
Cross-source data retrieval method, medium and device based on space-time asset catalogue Download PDFInfo
- Publication number
- CN117312688B CN117312688B CN202311612810.8A CN202311612810A CN117312688B CN 117312688 B CN117312688 B CN 117312688B CN 202311612810 A CN202311612810 A CN 202311612810A CN 117312688 B CN117312688 B CN 117312688B
- Authority
- CN
- China
- Prior art keywords
- space
- time
- data
- search
- conditions
- 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
- 238000000034 method Methods 0.000 title claims abstract description 55
- 239000013598 vector Substances 0.000 claims description 48
- 238000004590 computer program Methods 0.000 claims description 13
- 238000013507 mapping Methods 0.000 claims description 6
- 238000000926 separation method Methods 0.000 claims description 2
- 230000008569 process Effects 0.000 abstract description 7
- 230000010354 integration Effects 0.000 abstract description 6
- 230000009286 beneficial effect Effects 0.000 abstract description 3
- 230000001737 promoting effect Effects 0.000 abstract description 2
- 230000000875 corresponding effect Effects 0.000 description 33
- 238000004364 calculation method Methods 0.000 description 13
- 238000007726 management method Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 241000656416 Amazonia Species 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 239000000047 product Substances 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 101000880310 Homo sapiens SH3 and cysteine-rich domain-containing protein Proteins 0.000 description 1
- 102100037646 SH3 and cysteine-rich domain-containing protein Human genes 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 230000000153 supplemental effect Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/3332—Query translation
- G06F16/3338—Query expansion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9532—Query formulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9538—Presentation of query results
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a cross-source data retrieval method, medium and equipment based on a space-time asset directory, and belongs to the field of big data. Firstly, collecting massive heterogeneous space-time asset catalogues into a unified retrieval space, and carrying out word vectorization on semantic features of the heterogeneous space-time asset catalogues; then expanding the search condition based on the relevance among the conditions, searching in a search space by using the expanded search condition, and taking the searched space-time data as a preliminary search result; and finally grouping the space-time data in the preliminary search result according to the data source identification, and after each group of space-time data is respectively ordered according to the mixed association similarity, displaying the result as a final returned search result. The invention can realize high-efficiency retrieval and integration of massive cross-source space-time data, improves the flexibility and efficiency in the process of retrieving big data, and is beneficial to promoting the space-time data asset management and application in big data age.
Description
Technical Field
The invention relates to the field of big data, in particular to a method for realizing cross-source data retrieval in a massive space-time database consisting of different space-time asset catalogues.
Background
With the widespread use of Geographic Information Systems (GIS) and remote sensing technologies, more and more spatio-temporal data (e.g., geographic location and time information) are generated and accumulated, including satellite imagery, geographic data, sensor data, and the like. These spatio-temporal data come from different data sources, such as satellites, weather authorities, geographic information departments, etc., with different data formats, data structures, data content and data service interfaces. In practical applications, users need to acquire, integrate and fuse spatio-temporal data from multiple data sources to support various application scenarios, such as city planning, resource management, environmental protection, and the like. However, due to the heterogeneity between different data sources, the complexity of spatio-temporal data, and the complexity of data acquisition, there are still certain technical difficulties and challenges in implementing cross-source data retrieval.
The traditional cross-source data retrieval method is mostly dependent on complex processing procedures of manually written query sentences, data integration and fusion, and lacks of automation and intelligent capabilities, so that the problems of inaccurate retrieval results, difficult data integration, poor user experience and the like are caused. In addition, most of the existing data retrieval methods are aimed at a single data source, and no effective solution exists for cross-source retrieval of multi-source data. Therefore, there is a need for a new method and system that automatically retrieves spatiotemporal data from a plurality of data sources, providing a multi-source spatiotemporal data set with more accurate retrieval results, to meet the needs of users for multi-source spatiotemporal data.
Disclosure of Invention
The invention aims to solve the problems in the prior art and provides a cross-source data retrieval method based on a space-time asset directory.
In order to achieve the aim of the invention, the technical scheme adopted by the invention is as follows:
in a first aspect, the present invention provides a spatiotemporal asset directory-based cross-source data retrieval method, comprising:
s1, adding space-time asset catalogues from different data sources into a search space, traversing each piece of space-time data in each space-time asset catalogue in the search space, reading the data source identification and semantic feature information of each piece of space-time data from the space-time asset catalogue, and mapping the semantic feature information of each piece of space-time data into a semantic feature word vector through a word vector model;
s2, after receiving a search formula input by a user, analyzing all types of search conditions from the search formula, wherein the types of the search conditions comprise semantic feature conditions, time conditions, space conditions and resolution conditions; expanding the search conditions based on the relevance among the conditions for each original search condition analyzed from the search formula, searching in the search space by using the expanded search conditions, and taking the searched space-time data as a preliminary search result;
S3, grouping the space-time data in the preliminary search result according to the data source identification, and grouping the space-time data from the same data source into the same group; each group of space-time data is respectively sequenced according to the mixed association similarity and then is displayed as a final returned retrieval result; the mixed association similarity of each piece of space-time data is weighted by the semantic feature association degree, the time association degree, the space association degree and the resolution association degree between the piece of space-time data and the original retrieval condition.
As a preferable mode of the first aspect, in S1, the spatiotemporal data in the spatiotemporal asset directory added to the search space are remote sensing image data, and the semantic feature information of each piece of spatiotemporal data includes a semantic tag of the spatiotemporal data, a sensor for acquiring the spatiotemporal data, and a satellite on which the sensor is mounted.
Preferably, the semantic tag of the spatiotemporal data is composed of a title and description information of a spatiotemporal asset directory where the spatiotemporal data is located.
As a preferable mode of the first aspect, in S2, the method for expanding the search condition based on the correlation between conditions is as follows:
after analyzing the search expression, judging whether the analyzed search condition contains semantic feature conditions, time conditions, space conditions and resolution conditions;
If the search expression contains semantic feature conditions, performing similar word matching on each word in the semantic feature conditions in a corpus, and expanding a plurality of words with highest similarity obtained by matching each word in the corpus into semantic feature conditions during search;
if the search formula contains time conditions, taking the time conditions as central time, and taking the central time and expansion time periods respectively before and after the central time as the time conditions during search;
if the search formula contains space conditions, taking the corresponding space region range as a central space, and taking the central space or a larger space region range containing the central space as the space conditions during search;
if the search expression includes a resolution condition, the resolution level where the resolution condition is located is used as the resolution condition at the time of search according to a resolution level division standard preset for the space-time data.
As a preferable mode of the first aspect, if the spatial condition included in the search expression is a place name, the place name is mapped to the corresponding spatial region range coordinate according to the administrative division vector file.
As a preferred aspect of the first aspect, in S3, the method for calculating the mixed association similarity of each piece of spatio-temporal data includes:
Encoding semantic feature conditions analyzed from the search formula by adopting a word vector model to obtain an encoded first word vector, simultaneously taking a semantic feature word vector of a semantic feature information field corresponding to the semantic feature conditions in the current space-time data as a second word vector, and calculating the similarity of the first word vector and the second word vector as semantic feature association;
calculating a time crossing index and a time interval index between a first time span range corresponding to the time condition and a second time span range corresponding to the current space-time data for the time condition analyzed from the search formula, and taking the weighted sum of the time crossing index and the time interval index as a time association degree; the time crossing index is the degree of crossing between a first time span range and a second time span range, and the time interval index is the degree of range center interval between the first time span range and the second time span range;
calculating a space intersection index and a space distance index between a first space region range corresponding to the space condition and a second space region range corresponding to the current space-time data for the space condition analyzed from the search formula, and taking the weighted sum of the space intersection index and the space separation index as a space association degree; the space intersection index is the intersection degree between the first space region range and the second space region range, and the space distance index is the space region center interval degree between the first space region range and the second space region range;
Calculating absolute difference values of a first resolution corresponding to the resolution condition and a second resolution corresponding to the current space-time data for the resolution condition analyzed from the search formula, and dividing the absolute difference values by a resolution level where the resolution condition is located to obtain a quotient serving as a resolution association degree;
and carrying out weighted summation on the semantic feature association degree, the time association degree, the space association degree and the resolution association degree of the current space-time data to obtain the mixed association similarity between the space-time data and the original retrieval condition.
As a preferable aspect of the above first aspect, when calculating the time cross index, if the first time span range and the second time span range are the same or have an inclusion relationship, the time cross index is set to 1, if there is no intersection between the first time span range and the second time span range, the time cross index is set to 0, and if there is an incompletely included intersection relationship between the first time span range and the second time span range, the time cross index is set to a value within the (0, 1) interval;
when the time interval index is calculated, taking the sum of the spans of the first time span range and the second time span range as a numerator, taking the span of the first time span range, the span of the second time span range and the sum of the range center intervals of the two time span ranges as a denominator, and taking the quotient of the numerator and the denominator as the time interval index;
When the space intersection index is calculated, if the first space region range and the second space region range are the same or have an inclusion relationship, the space intersection index is set to be 1, if the first space region range and the second space region range are completely separated, the space intersection index is set to be 0, and if the first space region range and the second space region range have an incompletely included intersection relationship, the space intersection index is set to be a value in a (0, 1) interval;
and when the space distance index is calculated, mass center coordinates of the first space region range and the second space region range are respectively determined, and the distance between the two mass centers is used as the space distance index.
As a preferable aspect of the first aspect, the type of the search condition further includes a cloud amount condition for limiting cloud amount of the remote sensing image, and if the search condition analyzed from the search formula input by the user includes the cloud amount condition, the mixed association similarity is required to be multiplied by a cloud amount weight before being used for ranking; the cloud weight of each piece of space-time data is determined by the relative size of the cloud condition analyzed in the search formula and the cloud percentage of the piece of space-time data, if the cloud percentage of the space-time data is smaller than the upper limit of the cloud percentage corresponding to the cloud condition, the cloud weight is set to be 1, otherwise, the cloud weight is set to be a value in the interval of (0, 1) and the value is inversely related to the cloud percentage of the piece of space-time data.
In a second aspect, the present invention provides a computer readable storage medium, on which a computer program is stored, which when executed by a processor, implements a method for cross-source data retrieval based on spatiotemporal asset catalogs according to any of the above aspects.
In a third aspect, the present invention provides a computer electronic device comprising a memory and a processor;
the memory is used for storing a computer program;
the processor is configured to implement the spatiotemporal asset directory-based cross-source data retrieval method according to any of the above first aspects when executing the computer program.
Compared with the prior art, the invention has the beneficial effects that:
1) Aiming at the problem that the cross-source data is difficult to accurately retrieve, the invention realizes comprehensive and unified management of the data assets in different data sources by collecting the existing space-time asset catalogue, so as to realize high-efficiency cross-source data integration and retrieval in the same retrieval space.
2) The invention realizes the supplement of the search range by carrying out semantic association and space-time association on the search condition, and achieves accurate matching and associated recommendation according to the semantic tag and the space-time attribute of the data asset.
3) According to the method, the data asset ordering and retrieval optimization is realized by reordering based on the mixed association similarity of the space-time data, and the efficiency and the precision of cross-source data retrieval are improved. The mixed association similarity can be calculated according to the multi-dimensional factors of the data space-time data and optimized according to the retrieval requirements, so that the retrieval results meeting the user requirements are provided.
4) The invention can realize the high-efficiency retrieval and integration of cross-source data, improves the flexibility and efficiency of data retrieval, and is beneficial to promoting the space-time data asset management and application in big data age.
Drawings
FIG. 1 is a flow chart of steps of a method for cross-source data retrieval based on spatiotemporal asset inventory.
FIG. 2 is a schematic diagram of a hybrid associative similarity calculation in an embodiment.
FIG. 3 is a schematic diagram of a modified hybrid associative similarity calculation in another embodiment.
Detailed Description
In order that the above objects, features and advantages of the invention will be readily understood, a more particular description of the invention will be rendered by reference to the appended drawings. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. The present invention may be embodied in many other forms than described herein and similarly modified by those skilled in the art without departing from the spirit of the invention, whereby the invention is not limited to the specific embodiments disclosed below. The technical features of the embodiments of the invention can be combined correspondingly on the premise of no mutual conflict.
In the description of the present invention, it should be understood that the terms "first" and "second" are used solely for the purpose of distinguishing between the descriptions and not necessarily for the purpose of indicating or implying a relative importance or implicitly indicating the number of features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature.
Currently, various spatio-temporal data are provided by different data manufacturers and stored in different data sources, with different data structures, which makes cross-source data retrieval complex and difficult, requiring significant time and resources. Existing data retrieval methods often rely on specific data sources and data structures, limiting the flexibility and efficiency of data retrieval. The invention aims to provide a cross-source data retrieval method based on a space-time asset directory, which can effectively retrieve space-time data from different data manufacturers and data sources and realize efficient cross-source data integration and retrieval.
In a preferred embodiment of the present invention, a spatiotemporal asset directory-based cross-source data retrieval method is provided that can perform cross-source retrieval for all spatiotemporal data registered as meeting the spatiotemporal asset directory criteria. As shown in fig. 1, the specific implementation steps of the cross-source data retrieval method based on the space-time asset directory are as follows:
S1, adding space-time asset catalogues from different data sources into a search space, traversing each piece of space-time data in each space-time asset catalogue in the search space, reading the data source identification and semantic feature information of each piece of space-time data from the space-time asset catalogue, and mapping the semantic feature information of each piece of space-time data into a semantic feature word vector through a word vector model.
The invention can realize comprehensive and unified management of the data assets in different data sources by collecting all the existing space-time asset catalogs which can be searched into the search space. However, the spatiotemporal asset list added to the search space may be any spatiotemporal data satisfying the spatiotemporal asset list (SpatioTemporal Assets Catalogs, star) standard, and the typical data type is but not limited to the remote sensing image data acquired by the satellite-mounted sensor. The specific data standard of the space-time asset inventory can be found in the official manual (https:// stacspec. Org) and will not be described in detail. In general, a spatiotemporal asset directory includes information such as data source identification, spatiotemporal attributes (including temporal information and spatial information), semantic tags, sensors, onboard satellites, cloud cover, resolution, and the like.
In addition, it should be noted that the "data source identifier" in the present invention is the identification information for judging the unique source of the space-time asset directory, and the "semantic feature information" in the present invention includes the semantic information and the feature information of the space-time data, where the semantic feature information is composed of a series of information fields, and the data source identifier, the semantic information and the feature information of different space-time data may have differences, and need to be selected and determined according to the organization form of the space-time asset directory of the corresponding category. Taking remote sensing image data meeting the space-time asset catalogue standard as a typical example, the space-time data in the space-time asset catalogue are remote sensing image data, and semantic feature information of each piece of space-time data comprises a semantic tag of the space-time data, a sensor for acquiring the space-time data and a satellite for carrying the sensor. And semantic tags for spatiotemporal data may consist of the title and descriptive information of the spatiotemporal asset directory in which the spatiotemporal data resides. Further, according to the STAC standard, the data source identification of the remote sensing image can adopt the id information of the root directory level of the spatiotemporal asset directory, the id information corresponds to the title of the spatiotemporal asset directory, the semantic tag is the id information and description information of the root directory level of the spatiotemporal asset directory object, and the spatiotemporal attribute of the spatiotemporal data in the spatiotemporal asset directory is the spatial range and time information of the spatiotemporal data object. For example, CBERS/AMAZONIA on AWS is one of the remote sensing image data sources meeting the space-time asset directory standard, and is a CBERS/AMAZONIA remote sensing image with an AWS sensor, and the data source is identified as id information of a root directory level: the CBERS-AMAZONIA ROOT is characterized in that a certain remote sensing image under the data source is a search object, the space range is a polygon described in a GeoJSON format, the time information is 2021/8/19 UTC 2:59:29, the semantic tags are id information CBERS-AMAZONIA ROOT of a ROOT directory level and description information Catalogs of AMAZONIA and CBERS/4A problems' image on AWS of a space-time asset directory, a sensor for acquiring the space-time data is AWS, and a carrying satellite for carrying the sensor is CBERS.
In addition, it should be noted that, in order to improve efficiency, each information field needs to be encoded into a word vector in the word vector encoding process, so as to be matched with the search condition in the subsequent search process. However, since all fields of the semantic feature information are not necessarily input in the search condition, the word vectors corresponding to all the information fields are not required to participate in the subsequent semantic feature similarity calculation.
In the present invention, the specific form of the word vector model is not limited, and encoding can be performed. In the embodiment of the invention, a Word2Vec Word vector model is preferably constructed to realize Word vector coding, and is further applied to Word similarity calculation and semantic tag similarity calculation. word2vec is a simple neural network consisting of 1 input layer, 1 hidden layer and 1 output layer. The input layer uses one-hot coding. Namely: assuming n words, each word may be represented by an n-dimensional vector having only one position of 1 and the remaining positions of 0. The number of neural elements in the hidden layer represents the dimension of each word represented by a vector. For the weight matrix between the input layer and the hidden layer, its shape should be a matrix of [ vocab_size, hidden_size ]. The output layer is a vector of [ vocab_size ] size, each value representing the probability of outputting a word. In the Word2Vec model, there are mainly two training models of Skip-Gram and CBOW, and it is intuitively understood that Skip-Gram is a given current value to predict a context. While CBOW is a given context to predict the current value. In this embodiment, the pre-trained Word vector model preferably adopts a Word2Vec model based on Skip-gram, and the Word2Vec Word vector model can be constructed to encode the title, description information, the words such as the sensor and the carrying satellite in each semantic tag through the Skip-gram training corpus, so as to form the Word vector for calculating the similarity.
S2, after receiving a search formula input by a user, analyzing all types of search conditions from the search formula, wherein the types of the search conditions comprise semantic feature conditions, time conditions, space conditions and resolution conditions; and expanding the search conditions based on the relevance among the conditions for each original search condition analyzed from the search formula, searching in the search space by the expanded search conditions, and taking the searched space-time data as a preliminary search result.
It should be noted that, the type of the search condition included in the search expression input by the user is not completely fixed, and may be one or more of a semantic feature condition, a time condition, a space condition and a resolution condition, when there are multiple search conditions, the spatio-temporal data satisfying each search condition in the search space needs to be returned as a preliminary search result, and a specific search type analysis method and a search matching algorithm belong to the prior art in the search field, which are not repeated.
The semantic feature condition in the present invention also includes two dimensions, namely, semantic condition including spatiotemporal data, sensor condition for acquiring the spatiotemporal data, and satellite condition for carrying the sensor. However, the semantic feature condition input by the user does not necessarily need to include all-dimensional conditions, and may include only the semantic condition, but not the sensor condition and the satellite-mounted condition.
Because the search space of the invention is formed by space-time asset catalogues of different data sources, the difference of the data sources can cause the heterogeneity of space-time data among the different data sources in the search space, and the complexity of the space-time data itself exists, the cross-source data search of accurate matching is directly carried out according to the search condition, and the required search result is easy to be difficult to return. According to the invention, the search conditions are supplemented by carrying out semantic association and space-time association on the search conditions, and the search is carried out after the search range is expanded, so that the accurate matching and the associated recommendation are achieved according to the semantic tags and the space-time attributes of the data assets.
In the embodiment of the present invention, when the condition-based relevance expansion search condition is described above, it is necessary to perform relevance matching and condition expansion in different forms for different search condition types, and specifically, the method for the condition-based relevance expansion search condition may be implemented as follows:
firstly, after analyzing a search formula input by a user, judging whether the analyzed search condition comprises a semantic feature condition, a time condition, a space condition and a resolution condition, and then respectively carrying out corresponding association and expansion according to a) to d) according to the type of the contained search condition:
a) If the search expression input by the user contains semantic feature conditions, performing similar word matching on each word in the semantic feature conditions in the corpus, and expanding a plurality of words with highest similarity obtained by matching each word in the corpus into the semantic feature conditions during search.
It should be noted that, because the search expression input by the user may include one or more condition terms, each condition term needs to be matched with a similar term in the corpus, the number of the similar terms specifically matched with the search expression can be reasonably set according to the actual situation, and preferably 3 terms with the strongest semantic association degree are selected from the corpus to be expanded as the semantic feature condition during the search. For example, if the search condition includes a semantic feature condition, such as CBERS, word vectors of CBERS are calculated according to the Skip-gram-based Word2Vec model mentioned in the previous step, and 3 related words with the strongest semantic relevance are selected from the corpus as semantic search expansion conditions, and the method is not limited to using CBERS as the search conditions.
Thus, when searching, the original condition word and any expanded condition are satisfied, the semantic feature searching condition can be satisfied, and the searching range is expanded.
b) If the search expression input by the user includes a time condition, the time condition is taken as a center time, and the center time itself and the expansion time periods respectively before and after the center time are taken as the time condition during search.
The extended time condition is a time period before the center time and a time period after the center time, and the two time periods plus the center time itself form a time limit for the returned search result in search, so long as any one of the three time periods can be returned. However, the length of the extended time period can be reasonably set according to actual needs, and the method is not limited. In an embodiment of the present invention, the time search condition is extended according to the time length of the central time, and the time length is less than 1 month with the time scale of 1 month as a standard, so that the month and the front and back 2 months where the time information is located can be used as the supplementary time search condition; and if the time length is longer than 1 month, taking the time of 2 times before and after the time information as the supplementary time retrieval condition. For example, the search condition includes a time condition, such as 2018, 9, 1, and the time information is determined according to the type of the time information, and the length of the time information is less than the time scale of 1 month, so the time information is defined as a time point. Therefore, the search condition is that the search condition is supplemented from 1 st 8 th 2018 to 30 th 10 th 2018. If the time information is 2018, it is determined according to the type of the time information, and the length of the time information is greater than the time scale of 1 month, so the time information is defined as a time period. And thus 2017-2019 are supplemental time retrieval conditions.
c) If the search expression input by the user includes a spatial condition, the corresponding spatial region range is used as a central space, and the central space itself or a larger spatial region range including the central space is used as the spatial condition during search.
It should be noted that, when expanding the space conditions, the space information input by the user is often a place name in text form, for example, zhejiang province. At this time, it is necessary to first convert such spatial conditions into corresponding spatial region ranges. Therefore, if the space condition is a place name, the place name and the space range coordinate may be associated one by one based on the nationwide administrative division vector file, and the space geometric region obtained by projecting the vector file corresponding to the place name may be used as the space search condition. The purpose of the one-to-one correspondence of place names and space range coordinates is to correlate place name information with spatial position information in a space coordinate system. The mapping correspondence between the common place names and the space coordinates is constructed based on the standard place names, but some place names are common names, which are the same as the space coordinates represented by the standard place names, but have different expression modes, and the association relationship needs to be noted. Therefore, the place name information contained in the space information needs to be converted in advance between the place name colloquial name and the standard place name. Two methods for solving the conversion problem of the common names and the standard names are available, one method is to match the similarity, and on the basis of comparing a large number of corpora, the similarity correlation is carried out on the information of the names, and the common names of the names are matched with the names of administrative regions. The second method is to build a place name colloquial name table and extract the mapping relation between the place name colloquial names and the standard place names. Accordingly, the embodiment can convert the place name colloquially into the standard place name according to actual needs through the similarity matching or the pre-constructed mapping relation table.
In practical application, if the original spatial region range is small, the spatial region range can be further expanded outwards by taking the original spatial region range as the center, and a larger spatial region range containing the center space is taken as the spatial condition during searching. For example, the original search condition is Hangzhou city, west lake district, and the space region range of the whole Hangzhou city can be expanded to the space condition at the time of search.
d) If the search expression input by the user includes a resolution condition, the resolution level where the resolution condition is located is set as the resolution condition at the time of search according to a resolution level division standard preset for the spatio-temporal data.
It should be noted that, the specific resolution ranking criteria may be set according to actual situations. For example, in one embodiment of the present invention, the resolution dividing criteria may be classified into 3 levels, i.e., a high resolution image, a medium resolution image, and a low resolution image, if the resolution is in the order of meters, the high resolution image, if the resolution is in the order of ten meters, the medium resolution image, and if the resolution is in the order of hundred meters, the low resolution image. The resolution condition of the original input of the user is 50 meters, which belongs to the middle resolution image, so that the middle resolution image is taken as the resolution condition during retrieval, and any image sensing data belonging to the middle resolution image can be taken as the retrieval result and is not limited to the remote sensing image data of 50 meters.
Through the steps, the time information, the resolution information and the semantic feature information of cross-source retrieval are expanded, the place name retrieval range is optimized, the retrieval result can be more accurate through the accurate space range, and the association recommendation can be effectively carried out on the retrieval result through the expanded time information, resolution information and semantic feature information.
S3, grouping the space-time data in the preliminary search result according to the data source identification, and grouping the space-time data from the same data source into the same group; each group of space-time data is respectively sequenced according to the mixed association similarity and then is displayed as a final returned retrieval result; wherein each piece of spatiotemporal data is mixed and correlated with similarityThe semantic feature correlation degree, the time correlation degree, the space correlation degree and the resolution correlation degree between the piece of space-time data and the original retrieval condition are weighted to obtain.
In the invention, because a plurality of data sources are involved, in order to avoid the possibility of similar or repeated codes possibly existing among the plurality of data sources, the space-time data in the preliminary retrieval result are grouped according to the data source identifiers, thereby meeting the aim of orderly displaying the grouping, and the retrieval result is respectively displayed through the data source identifiers, so that the confusion of the space-time data of different sources in the sorting process is avoided. According to the sorting process, the weight calculation is needed to be carried out on the search result based on the semantic relevance, the time-space relevance, the data product grade and other factors, so that the data asset sorting and the search optimization are realized, and the accuracy of cross-source data search is improved. The specific weight evaluation can be calculated according to factors such as space-time attributes and semantic association of the data assets, and is optimized according to the retrieval requirements, so that the retrieval results meeting the user requirements are provided.
In the embodiment of the invention, the method for calculating the mixed association similarity of each piece of space-time data is shown as S31-S35, and specifically comprises the following steps:
s31, coding semantic feature conditions analyzed from the search formula by adopting a word vector model to obtain a coded first word vector, simultaneously taking a semantic feature word vector of a semantic feature information field corresponding to the semantic feature conditions in the current space-time data as a second word vector, and calculating similarity of the first word vector and the second word vector as semantic feature association degree.
In the step S1, each information field in the semantic feature information is word vector encoded, but the semantic feature condition analyzed in the search expression does not necessarily include all the corresponding information fields, so that the information field constructing the second word vector is required to be identical to the information field included in the semantic feature condition analyzed in the search expression when calculating the semantic feature relevance. For example, the semantic feature conditions analyzed from the search expression only include the semantic condition and the sensor condition, but the carried satellite condition is absent, so that the information field of the second word vector only needs to adopt the semantic tag field of the current spatiotemporal data and the word vector corresponding to the carried satellite field.
In addition, in practical application, taking remote sensing image data as an example, id information and description information of a root directory level of each spatiotemporal data asset directory are a semantic information set, and the extracted semantic information can reflect the subject content represented by the data and is used for matching the content of a data source of the spatiotemporal data asset directory. The computation of semantic feature relevance therefore requires the computation of information field similarity within both sets. The total semantic feature relevance between the semantic feature information of the search result and the semantic feature information of the search condition can be determined by pairing the semantic feature information fields with the highest degree of mutual relevance and calculating the average degree of mutual relevance. And using the extracted semantic information set to measure and calculate the topic similarity between the search result and the search condition, and loading the trained word vector model by using a Gensim library. Screening each pair of information fields with highest matching degree between two semantic feature information setsAnd after pairing is completed, calculating the semantic feature association degree of each pair of semantic feature information, and obtaining the average similarity. For example, the first semantic feature information set a contains M information field words, the second semantic feature information set B contains N information field words (M may be equal to N or not), then for each information field word in a, find the information field word with the highest word vector matching degree in B, and pair the two to form a group of word pairs; for a total M groups of word pairs, calculating the average similarity of word vectors, and obtaining average similarity, namely the semantic feature association degree of the space-time data object in the space-time asset catalogue and the retrieval condition 。
S32, calculating a time crossing index between a first time span range corresponding to the time condition and a second time span range corresponding to the current space-time data for the time condition analyzed from the search formulaAnd time interval index>And time Cross index +.>And time interval index>Weighted summation as time correlation +.>。
It should be noted that, in the embodiment of the present invention, when calculating the time intersection index, if the first time span range and the second time span range are the same or have an inclusion relationship, the time intersection index is set to 1, if there is no intersection between the first time span range and the second time span range, the time intersection index is set to 0, and if there is an incompletely included intersection relationship between the first time span range and the second time span range, the time intersection index is set to a value within the (0, 1) interval.
In an exemplary embodiment, a time cross index of time information of the search result and a time condition of the search condition is calculatedAnd time interval index >When the time types of the first time span range and the second time span range can be defined according to the time lengths: taking a time scale of 1 month as a standard, defining a time point when the time length of the first time span range and the second time span range is less than 1 month; the time length of the first time span range and the second time span range is greater than 1 month, defined as a time period. Time Cross index->Is calculated according to the following rule:
(1) If the time point and the time point are the same in year and month, namely the time points are considered to be the same, the time crossing index is 1;
(2) If the time point and the time point are the same in the year but different in the month or different in the year and month, the time point and the time point are regarded as different in the time point, and the time crossing index is 0;
(3) If the time period comprises a time point, the two time span ranges have a containing relation, and the time crossing index is 1;
(4) If the time period does not contain the time point, the two time span ranges do not have a containing relation, and the time crossing index is 0;
(5) If a time period includes or equals another time period, the time crossing index is 1;
(6) If a time period crosses another time period but does not completely contain, the time crossing index is 0.5;
(7) If a time period is completely separated from another time period, the time crossing index is 0.
Likewise, the above time interval indexIs the extent of the range center interval between the first time span range and the second time span range. In the embodiment of the invention, when calculating the time interval index, the sum of the spans of the first time span range and the second time span range is taken as a numerator, the span of the first time span range, the span of the second time span range and the sum of the range center intervals of the two time span ranges are taken as a denominator, and the quotient of the numerator and the denominator is taken as the time interval index.
In an exemplary embodiment, the time interval indexThe calculation of (1) adopts the following calculation formula:
wherein,for a time span of the first time span range, +.>For the time span of the second time span range,is the duration of the interval between the center of the range of the first time span range and the center of the range of the second time span range.
Thus, the time correlation degreeThe calculation formula of (2) is as follows:
wherein:and->All are weight values, and are optimizable super parameters.
S33, calculating a space intersection index between a first space region range corresponding to the space condition and a second space region range corresponding to the current space-time data for the space condition analyzed from the search formula And spatial distance index>And spatial cross index +.>And spatial distance index>Weighted summation as spatial correlation +.>。
It should be noted that the spatial cross index is the degree of cross between the first spatial region range and the second spatial region range. In the embodiment of the present invention, when calculating the above-mentioned spatial cross index, if the first spatial region range and the second spatial region range are the same or have an inclusion relationship, the spatial cross index is set to 1, if the first spatial region range and the second spatial region range are completely separated, the spatial cross index is set to 0, and if the first spatial region range and the second spatial region range have an incompletely included intersection relationship, the spatial cross index is set to a value within the (0, 1) interval.
The above spatial cross indexThe method can be used for measuring the relation between the space information of the search condition and the space information of the search result, and carrying out topological relation measurement on the mapped space geometry, so that the space association degree of the space information of the search condition and the space information of the search result can be obtained. In an exemplary embodiment, the spatial intersection index of the first spatial region range and the second spatial region range may be determined according to the following rule based on a nine-intersection model >：
(1) If the two space regions are equal in scope or have inclusion relation, the space intersection index is 1;
(2) If the two space regions have an intersection part but are not completely equal, or the boundaries are connected, the space intersection index is 0.5;
(3) If the two spatial regions are completely separated, the spatial cross index is 0.
Similarly, the spatial distance index is a degree of spatial region center interval between the first spatial region range and the second spatial region range. In an embodiment of the present invention, when calculating the above spatial distance index, the centroid coordinates of the first spatial region range and the second spatial region range may be determined first, and the distance between the two centroids is used as the spatial distance indexThe calculation formula is as follows:
wherein:for the centroid coordinates of the first spatial region range, +.>Is the centroid coordinates of the second spatial region range.
Thereby, the above spatial correlation degreeThe calculation formula of (2) is as follows:
wherein:and->All are weight values, and are optimizable super parameters.
S34, calculating absolute difference values of a first resolution corresponding to the resolution condition and a second resolution corresponding to the current space-time data for the resolution condition analyzed from the search formula, and dividing the absolute difference values by a resolution level where the resolution condition is located to obtain a quotient serving as a resolution association degree.
In the foregoing embodiment of the present invention, the resolution is divided into three levels, namely, a high resolution image if the resolution is in the order of meters, a medium resolution image if the resolution is in the order of ten meters, and a low resolution image if the resolution is in the order of hundred meters. Thus, if the first resolution corresponding to the resolution condition is recorded asThe second resolution of the current spatio-temporal data itself is noted +.>，/>The resolution level at which this is recorded is +.>Corresponding resolution relevance->Computing means of (a)The formula is as follows:
wherein: resolution levelAssignment is required, if the resolution level is high resolution image, +.>1, if the resolution level is a middle resolution image, then +.>10, if the resolution level is low resolution image>100.
And S35, carrying out weighted summation on the semantic feature association degree, the time association degree, the space association degree and the resolution association degree of the current space-time data to obtain the mixed association similarity between the space-time data and the original retrieval condition.
As shown in FIG. 2, when the above-mentioned degree of association by semantic features is obtainedTime relevance->Spatial association->Resolution relevance- >Weighted summation is carried out to calculate the mixed association similarity +.>The calculation formula of (2) is expressed as follows:
wherein:association degree for semantic features->Weight of->For the time association +.>Weight of->For spatial association +.>Weight of->For resolution relevance->Is a weight of (2).
When the weighted summation obtains the mixed association similarity of each search resultThen, for a group of spatiotemporal data from the same data source, the similarity can be associated according to the mixture>And sequencing, and displaying the set of space-time data from high to low according to the mixed association similarity. Because of the existence of a plurality of groups of space-time data retrieval results with different sources, each group can be arranged in a similarity descending order according to the mode, and then displayed in groups for a user to check.
In the present invention, for the specific spatiotemporal data of the remote sensing image, the cloud amount when the remote sensing image is acquired has an important image for the quality of the remote sensing image, so the data quality based on the cloud amount also needs to be considered when the remote sensing image is searched. Thus, in another embodiment of the present invention, the type of the search condition may further include a cloud amount condition that limits the cloud amount of the remote sensing image, i.e. the user may input the allowable cloud amount, and the cloud amount information is typically a percentage, for example, the user may input 10%, i.e. search the remote sensing image with the cloud amount not exceeding 10%. Therefore, as shown in fig. 3, if the search condition analyzed from the search expression input by the user includes the cloud amount condition, the cloud amount weight needs to be multiplied before the mixed association similarity is used for ranking, that is, the mixed association similarity calculation formula may be modified into the following form:
Wherein:is cloud weight.
In the embodiment of the invention, the cloud weight of each piece of space-time data is determined by the relative size of the cloud condition analyzed in the search formula and the cloud percentage of the space-time data, and if the cloud percentage of the space-time data is smaller than the upper limit of the cloud percentage corresponding to the cloud condition, the cloud weight is determinedSet to 1, otherwise the cloud weight is +.>The value in the interval (0, 1) is set and is inversely related to the cloud percentage of the space-time data. In an exemplary implementation, if the cloud percentage of the spatiotemporal data itself in the search result +.>The cloud weight is set to be 1 when the cloud weight is smaller than the upper limit of the cloud percentage corresponding to the cloud condition; if the cloud percentage of the space-time data per se in the search result is +.>When the cloud weight is larger than the upper limit of the percentage of the cloud weight corresponding to the cloud weight condition, the cloud weight is set as +.>。
In addition, the cross-source data retrieval method based on the space-time asset directory in the above embodiment may be essentially executed by a computer program. Thus, also, based on the same inventive concept, there is also provided in another preferred embodiment of the present invention a computer electronic device corresponding to the method provided in the above embodiment, including a memory and a processor;
The memory is used for storing a computer program;
the processor is configured to implement the spatiotemporal asset directory-based cross-source data retrieval method in the above embodiment when executing the computer program.
Further, the logic instructions in the memory described above may be implemented in the form of software functional units and stored in a computer-readable storage medium when sold or used as a stand-alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention.
Thus, based on the same inventive concept, there is also provided in another preferred embodiment of the present invention a computer readable storage medium corresponding to the method provided in the above embodiment, where the storage medium stores a computer program, and when the computer program is executed by a processor, the method for searching cross-source data based on a space-time asset directory in the above embodiment can be implemented.
Specifically, in the computer-readable storage medium of the above two embodiments, the stored computer program is executed by the processor, and the steps S1 to S3 may be executed.
It is understood that the storage medium may include random access Memory (Random Access Memory, RAM) or may include Non-Volatile Memory (NVM), such as at least one magnetic disk Memory. Meanwhile, the storage medium may be various media capable of storing program codes, such as a USB flash disk, a mobile hard disk, a magnetic disk or an optical disk.
It will be appreciated that the above-described processor may be a general-purpose processor, including a central processing unit (Central Processing Unit, CPU), a network processor (Network Processor, NP), etc.; but also digital signal processors (Digital Signal Processing, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), field programmable gate arrays (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.
It should be further noted that, for convenience and brevity of description, specific working processes of the system described above may refer to corresponding processes in the foregoing method embodiments, which are not described herein again. In the embodiments provided in the present application, the division of steps or modules in the system and the method is merely one logic function division, and there may be another division manner when actually implemented, for example, a plurality of modules or steps may be combined or may be integrated together, and one module or step may also be split.
The above embodiment is only a preferred embodiment of the present invention, but it is not intended to limit the present invention. Various changes and modifications may be made by one of ordinary skill in the pertinent art without departing from the spirit and scope of the present invention. Therefore, all the technical schemes obtained by adopting the equivalent substitution or equivalent transformation are within the protection scope of the invention.
Claims (10)
1. A cross-source data retrieval method based on space-time asset catalogs is characterized by comprising the following steps:
s1, adding space-time asset catalogues from different data sources into a search space, traversing each piece of space-time data in each space-time asset catalogue in the search space, reading the data source identification and semantic feature information of each piece of space-time data from the space-time asset catalogue, and mapping the semantic feature information of each piece of space-time data into a semantic feature word vector through a word vector model;
s2, after receiving a search formula input by a user, analyzing all types of search conditions from the search formula, wherein the types of the search conditions comprise semantic feature conditions, time conditions, space conditions and resolution conditions; expanding the search conditions based on the relevance among the conditions for each original search condition analyzed from the search formula, searching in the search space by using the expanded search conditions, and taking the searched space-time data as a preliminary search result;
S3, grouping the space-time data in the preliminary search result according to the data source identification, and grouping the space-time data from the same data source into the same group; each group of space-time data is respectively sequenced according to the mixed association similarity and then is displayed as a final returned retrieval result; the mixed association similarity of each piece of space-time data is weighted by the semantic feature association degree, the time association degree, the space association degree and the resolution association degree between the piece of space-time data and the original retrieval condition.
2. The method for cross-source data retrieval based on spatiotemporal asset catalogue according to claim 1, wherein in S1, spatiotemporal data in the spatiotemporal asset catalogue added into the retrieval space are remote sensing image data, and semantic feature information of each piece of spatiotemporal data comprises semantic tags of the spatiotemporal data, a sensor for acquiring the spatiotemporal data and a satellite carrying the sensor.
3. The method for cross-source data retrieval based on spatiotemporal asset catalogs according to claim 2, wherein the semantic tags of the spatiotemporal data consist of the title and description information of the spatiotemporal asset catalogs where the spatiotemporal data are located.
4. The method for cross-source data retrieval based on space-time asset catalogue according to claim 2, wherein in S2, the method for expanding the retrieval condition based on the correlation between conditions is as follows:
After analyzing the search expression, judging whether the analyzed search condition contains semantic feature conditions, time conditions, space conditions and resolution conditions;
if the search expression contains semantic feature conditions, performing similar word matching on each word in the semantic feature conditions in a corpus, and expanding a plurality of words with highest similarity obtained by matching each word in the corpus into semantic feature conditions during search;
if the search formula contains time conditions, taking the time conditions as central time, and taking the central time and expansion time periods respectively before and after the central time as the time conditions during search;
if the search formula contains space conditions, taking the corresponding space region range as a central space, and taking the central space or a larger space region range containing the central space as the space conditions during search;
if the search expression includes a resolution condition, the resolution level where the resolution condition is located is used as the resolution condition at the time of search according to a resolution level division standard preset for the space-time data.
5. The method for cross-source data retrieval based on space-time asset inventory according to claim 4, wherein if the spatial condition included in the retrieval formula is a place name, the place name is mapped to the corresponding spatial region range coordinates according to an administrative division vector file.
6. The method for cross-source data retrieval based on space-time asset catalogue according to claim 2, wherein in S3, the method for calculating the mixed association similarity of each piece of space-time data is as follows:
encoding semantic feature conditions analyzed from the search formula by adopting a word vector model to obtain an encoded first word vector, simultaneously taking a semantic feature word vector of a semantic feature information field corresponding to the semantic feature conditions in the current space-time data as a second word vector, and calculating the similarity of the first word vector and the second word vector as semantic feature association;
calculating a time crossing index and a time interval index between a first time span range corresponding to the time condition and a second time span range corresponding to the current space-time data for the time condition analyzed from the search formula, and taking the weighted sum of the time crossing index and the time interval index as a time association degree; the time crossing index is the degree of crossing between a first time span range and a second time span range, and the time interval index is the degree of range center interval between the first time span range and the second time span range;
calculating a space intersection index and a space distance index between a first space region range corresponding to the space condition and a second space region range corresponding to the current space-time data for the space condition analyzed from the search formula, and taking the weighted sum of the space intersection index and the space separation index as a space association degree; the space intersection index is the intersection degree between the first space region range and the second space region range, and the space distance index is the space region center interval degree between the first space region range and the second space region range;
Calculating absolute difference values of a first resolution corresponding to the resolution condition and a second resolution corresponding to the current space-time data for the resolution condition analyzed from the search formula, and dividing the absolute difference values by a resolution level where the resolution condition is located to obtain a quotient serving as a resolution association degree;
and carrying out weighted summation on the semantic feature association degree, the time association degree, the space association degree and the resolution association degree of the current space-time data to obtain the mixed association similarity between the space-time data and the original retrieval condition.
7. The method for searching cross-source data based on space-time asset catalogue according to claim 6, wherein when calculating the time crossing index, if the first time span range and the second time span range are the same or have an inclusion relationship, the time crossing index is set to 1, if there is no intersection between the first time span range and the second time span range, the time crossing index is set to 0, and if there is an incompletely included intersection relationship between the first time span range and the second time span range, the time crossing index is set to a value within a (0, 1) interval;
when the time interval index is calculated, taking the sum of the spans of the first time span range and the second time span range as a numerator, taking the span of the first time span range, the span of the second time span range and the sum of the range center intervals of the two time span ranges as a denominator, and taking the quotient of the numerator and the denominator as the time interval index;
When the space intersection index is calculated, if the first space region range and the second space region range are the same or have an inclusion relationship, the space intersection index is set to be 1, if the first space region range and the second space region range are completely separated, the space intersection index is set to be 0, and if the first space region range and the second space region range have an incompletely included intersection relationship, the space intersection index is set to be a value in a (0, 1) interval;
and when the space distance index is calculated, mass center coordinates of the first space region range and the second space region range are respectively determined, and the distance between the two mass centers is used as the space distance index.
8. The method for searching cross-source data based on space-time asset catalogue according to claim 6, wherein the types of the search conditions further comprise cloud amount conditions for limiting cloud amount of remote sensing images, and if the search conditions analyzed from the search formula input by the user contain cloud amount conditions, the cloud amount weights need to be multiplied before the mixed association similarity is used for sorting; the cloud weight of each piece of space-time data is determined by the relative size of the cloud condition analyzed in the search formula and the cloud percentage of the piece of space-time data, if the cloud percentage of the space-time data is smaller than the upper limit of the cloud percentage corresponding to the cloud condition, the cloud weight is set to be 1, otherwise, the cloud weight is set to be a value in the interval of (0, 1) and the value is inversely related to the cloud percentage of the piece of space-time data.
9. A computer readable storage medium, wherein a computer program is stored on the storage medium, and when the computer program is executed by a processor, the method for searching cross-source data based on space-time asset catalogue according to any one of claims 1-8 is implemented.
10. A computer electronic device comprising a memory and a processor;
the memory is used for storing a computer program;
the processor is configured to implement the spatiotemporal asset directory-based cross-source data retrieval method according to any one of claims 1 to 8 when executing the computer program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311612810.8A CN117312688B (en) | 2023-11-29 | 2023-11-29 | Cross-source data retrieval method, medium and device based on space-time asset catalogue |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311612810.8A CN117312688B (en) | 2023-11-29 | 2023-11-29 | Cross-source data retrieval method, medium and device based on space-time asset catalogue |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117312688A CN117312688A (en) | 2023-12-29 |
| CN117312688B true CN117312688B (en) | 2024-01-26 |
Family
ID=89285221
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311612810.8A Active CN117312688B (en) | 2023-11-29 | 2023-11-29 | Cross-source data retrieval method, medium and device based on space-time asset catalogue |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117312688B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117688053A (en) * | 2024-02-02 | 2024-03-12 | 中国科学院空天信息创新研究院 | Remote sensing data retrieval method and device combined with land type and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022088409A1 (en) * | 2020-10-28 | 2022-05-05 | 中国商用飞机有限责任公司北京民用飞机技术研究中心 | Interactive retrieval method and apparatus, and computer device and storage medium |
| CN114676206A (en) * | 2022-03-31 | 2022-06-28 | 中国计量大学 | Unified organization expression and retrieval method for time-space elements of multi-element heterogeneous silk heritage |
| WO2023034358A2 (en) * | 2021-09-01 | 2023-03-09 | Graphika Technologies, Inc. | Analyzing social media data to identify markers of coordinated movements, using stance detection, and using clustering techniques |
| WO2023065617A1 (en) * | 2021-10-21 | 2023-04-27 | 北京邮电大学 | Cross-modal retrieval system and method based on pre-training model and recall and ranking |
| CN117076652A (en) * | 2023-10-17 | 2023-11-17 | 天启黑马信息科技(北京)有限公司 | Semantic text retrieval method, system and storage medium for middle phrases |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11392987B2 (en) * | 2013-10-09 | 2022-07-19 | Mobile Technology Corporation | Systems and methods for using spatial and temporal analysis to associate data sources with mobile devices |
| US20230252224A1 (en) * | 2021-01-22 | 2023-08-10 | Bao Tran | Systems and methods for machine content generation |
| US11748555B2 (en) * | 2021-01-22 | 2023-09-05 | Bao Tran | Systems and methods for machine content generation |
-
2023
- 2023-11-29 CN CN202311612810.8A patent/CN117312688B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022088409A1 (en) * | 2020-10-28 | 2022-05-05 | 中国商用飞机有限责任公司北京民用飞机技术研究中心 | Interactive retrieval method and apparatus, and computer device and storage medium |
| WO2023034358A2 (en) * | 2021-09-01 | 2023-03-09 | Graphika Technologies, Inc. | Analyzing social media data to identify markers of coordinated movements, using stance detection, and using clustering techniques |
| WO2023065617A1 (en) * | 2021-10-21 | 2023-04-27 | 北京邮电大学 | Cross-modal retrieval system and method based on pre-training model and recall and ranking |
| CN114676206A (en) * | 2022-03-31 | 2022-06-28 | 中国计量大学 | Unified organization expression and retrieval method for time-space elements of multi-element heterogeneous silk heritage |
| CN117076652A (en) * | 2023-10-17 | 2023-11-17 | 天启黑马信息科技(北京)有限公司 | Semantic text retrieval method, system and storage medium for middle phrases |
Non-Patent Citations (2)
| Title |
|---|
| 国外语义网研究现状与动向――基于2002―2018年ISWC会议;许鑫;杨佳颖;;情报学报(07);全文 * |
| 基于时空大数据的多规融合分析平台;文莉莉;;电子元器件与信息技术(04);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117312688A (en) | 2023-12-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101350012B (en) | Method and system for matching address | |
| US9569499B2 (en) | Method and apparatus for recommending content on the internet by evaluating users having similar preference tendencies | |
| CN117312688B (en) | Cross-source data retrieval method, medium and device based on space-time asset catalogue | |
| CN106503223B (en) | online house source searching method and device combining position and keyword information | |
| CN110162754B (en) | Method and equipment for generating post description document | |
| KR20210082112A (en) | An apparatus and a method for dual learning of low-variablility time series data based on virtual data | |
| CN113868235A (en) | Big data-based information retrieval and analysis system | |
| CN102081666B (en) | Index construction method and device for distributed picture search | |
| CN111159563A (en) | Method, device and equipment for determining user interest point information and storage medium | |
| CN113468034A (en) | Data quality evaluation method and device, storage medium and electronic equipment | |
| CN112330227A (en) | Urban management capability evaluation method and system based on digital urban management service construction | |
| CN112100506B (en) | Information pushing method, system, equipment and storage medium | |
| CN114049172A (en) | Item recommendation sorting method and device, computer equipment and storage medium | |
| CN113935401A (en) | Article information processing method, article information processing device, article information processing server and storage medium | |
| US20120310938A1 (en) | Information organizing sytem and information organizing method | |
| CN112579816A (en) | Remote sensing image retrieval method and device, electronic equipment and storage medium | |
| KR101823463B1 (en) | Apparatus for providing researcher searching service and method thereof | |
| CN113360586A (en) | Address aggregation degree query method, device and equipment and computer readable storage medium | |
| CN111581245B (en) | Data searching method and device | |
| CN115796984A (en) | Training method of item recommendation model, storage medium and related equipment | |
| CN115204436A (en) | Method, device, equipment and medium for detecting abnormal reasons of business indexes | |
| CN111125556A (en) | Recommendation sorting method and system, electronic device and readable storage medium | |
| CN117271577B (en) | Keyword retrieval method based on intelligent analysis | |
| CN111198912A (en) | Address data processing method and device | |
| CN116431711B (en) | Data intelligent acquisition method and system based on data characteristics |
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 |