CN114969350A - Intelligent auxiliary generation method for comprehensive situation - Google Patents
Intelligent auxiliary generation method for comprehensive situation Download PDFInfo
- Publication number
- CN114969350A CN114969350A CN202210914211.0A CN202210914211A CN114969350A CN 114969350 A CN114969350 A CN 114969350A CN 202210914211 A CN202210914211 A CN 202210914211A CN 114969350 A CN114969350 A CN 114969350A
- Authority
- CN
- China
- Prior art keywords
- target
- situation
- association
- information
- time
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- 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
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/10—Text processing
- G06F40/194—Calculation of difference between files
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides an intelligent auxiliary generation method for comprehensive situation, which comprises the following steps: the comprehensive environment construction module presents the geographic environment, the electromagnetic environment, the natural resources and the human geography in a characteristic vector form and superposes the geographic environment, the electromagnetic environment, the natural resources and the human geography on the digital earth layer by layer to form a multilayer superposed comprehensive environment; the target situation construction module generates target situation information from the regional real-time target data; the data association and integration module is based on a multi-layer overlapped integrated environment and combines target situation information, intelligent generation of the target situation is achieved through the modes of entity extraction, theme association, space-time mapping and semantic association, the target situation is associated with dynamic information, a target system and major events, and finally a full-dimensional multi-domain integrated situation product is formed. The invention enables the situation elements to serve the information users more objectively and comprehensively through the information data.
Description
Technical Field
The invention relates to the technical field of big data analysis and natural language processing, in particular to an intelligent auxiliary comprehensive situation generation method.
Background
The situation information is an important component in information security, and the safety situation can be analyzed and judged through the situation, so that the situation information is more important comprehensively and objectively. The construction and generation of situation information covering multiple dimensions become a basic functional task in information assurance, and are key indexes for evaluating the situation information quality level. The multidimensional situation information is used as an information data set with extremely strong comprehensiveness and extremely high complexity, and is not simple integration of information data of different types and different contents, but accurate systematic organization of multi-source, multi-content and multi-type information data. The system is essentially systematic information data, and comprises various target states, target basic information, comprehensive environment, global major events and other multi-type information characteristics.
The current situation map mainly takes real-time states such as target position, track and the like, the contents of historical conditions, dynamic rules, trend prediction and the like are relatively lacked, and the requirements of information guarantee on increasing development are not well met, so how to further expand and present situation elements to enable the situation elements to serve information users more objectively and comprehensively is an important problem which needs to be solved urgently.
Disclosure of Invention
The invention aims to provide an intelligent auxiliary generation method for comprehensive situation, which is used for further expanding and presenting situation elements, so that the situation elements can be more objectively and comprehensively served to information users through information data.
The invention provides an intelligent auxiliary generation method of comprehensive situation, which comprises the following steps:
s100, a comprehensive environment construction module presents the geographic environment, the electromagnetic environment, the natural resources and the humanistic geography in a characteristic vector form and superposes the geographic environment, the electromagnetic environment, the natural resources and the humanistic geography on the digital earth layer by layer to form a multilayer superposed comprehensive environment;
s200, a target situation construction module generates target situation feature vectors from the regional real-time target data;
s300, the data association and synthesis module depends on a multi-layer overlapped synthetic environment and combines target situation characteristic vectors to realize intelligent generation of target situations and association of regional target situations to information, a target system and major events through the ways of entity extraction, topic association, space-time mapping and semantic association, and finally comprehensive situation information of full-dimensional multi-domain is formed.
In some embodiments, the method for presenting the geographic environment in the form of the feature vector in step S100 is:
constructing geographic environment feature vectors of various countries according to the geographic environment data, wherein the elements of the geographic environment feature vectors comprise important ports, important organizations and strategic important roads; thus, the geographic environment feature vector is represented by the form:
wherein the content of the first and second substances,
is as followsiThe country is in the agetThe feature vector of the geographical environment of (a),tin the course of the time of the year,Nis the number of countries;
is as followsiThe country is in the agetThe important port distribution characteristic vector;
is as followsiThe country is in the agetThe important organizational structure distribution feature vector;
are respectively the firstiThe country is in the agetThe strategic key path of (1) distributes feature vectors.
In some embodiments, the method for presenting the electromagnetic environment in the form of a feature vector in step S100 is:
constructing an electromagnetic environment characteristic vector according to electromagnetic environment data, wherein elements of the electromagnetic environment characteristic vector comprise a channel attenuation coefficient, an electromagnetic interference coefficient, anti-interference capability and spectrum utilization rate; whereby the electromagnetic environment feature vector is represented in the form:
wherein the content of the first and second substances,
is as followspIs in the first areattThe feature vector of the electromagnetic environment at the time of day,
is as followspIs in the first areattThe channel attenuation coefficient characteristics at the time of day,
is as followspIs in the first areattThe characteristic of the electromagnetic interference coefficient at the moment,
is as followspIs in the first areattThe characteristics of the anti-interference capability at the moment,
is as followspIn the first areattSpectral usage characteristics at a time.
In some embodiments, the method for presenting the natural resources in the form of feature vectors in step S100 is as follows:
constructing a natural resource feature vector according to the natural resource data, wherein elements of the natural resource feature vector comprise key mineral products and energy; thus, the natural resource feature vector is represented in the form:
wherein the content of the first and second substances,
is a firstiThe country is in the agetThe natural resource feature vector of (2);
、
is as followsiThe country is in the agetThe key mineral distribution eigenvectors and the energy distribution eigenvectors.
In some embodiments, the method for presenting the human geography in the form of the feature vector in step S100 is as follows:
constructing a human geographical feature vector according to human geographical data, wherein elements of the human geographical feature vector comprise population number, religion number and average education year number; thus, the human geographic feature vector is represented in the form of:
wherein the content of the first and second substances,tin the course of the time of the year,
is as followsiThe country is in the yeartThe human-geographic feature vector of (a),Nthe number of countries is the number of countries,
is as followsiThe country is in the agetThe number of population of (a) is,
is a firstiThe country is in the agetThe elements of the religious quantity of (c),
is as followsiThe country is in the agetAverage educational age factor, [ … ]] T Is a transposition of a vector；
Combining the humanistic geographic feature vectors of all countries to form a world humanistic geographic feature matrix, wherein the expression form is as follows:
wherein the content of the first and second substances,tin the course of the time of the year,
is a world humanistic geographic feature matrix,
is a firstiThe country is in the agetThe human geographic feature vector of (1); elements in the world humanity geographic feature matrix are stored in a database table form.
In some embodiments, the method for the target situation construction module to generate the target situation feature vector from the regional real-time target data in step S200 is as follows:
the dynamic elements of the target situation feature vector comprise a target type, longitude, latitude, altitude and speed; therefore, the target situation feature vector is expressed in the form of:
wherein the content of the first and second substances,
is as followsiiThe object is atttTarget situation feature vectors of moments;
is shown asiiThe object is atttA target type of time;
is shown asiiAn object is atttA target longitude of the time of day;
denotes the firstiiThe object is atttTarget dimensions of time;
is shown asiiThe object is atttA target height at a time;
is shown asiiThe object is atttThe target speed at the moment.
In some embodiments, step S300 includes the following sub-steps:
s301, generating a target situation:
firstly, a target space state is presented on the digital earth in a target eigenvector form on the basis of the digital earth and a multi-layer superposed comprehensive environment, elements of the target eigenvector comprise space-time attributes, target countries, target names, target models and target membership, and the target eigenvector is represented in the form:
wherein the content of the first and second substances,
is as followsiiThe object is atttTarget feature vectors of the time;
is as followsiiThe object is atttThe target situation characteristic vector of the moment comprises space-time attributes and target model elements;
is used as the target membership element and is provided with a plurality of target membership elements,
is a target name element，
Is a target country element;
secondly, the situation is associated with a target knowledge base, similarity calculation is carried out according to the target name in the target characteristic vector and the target name in the knowledge base, information of the relevant target knowledge base is associated based on the similarity calculation result, and the calculation formula is as follows:
wherein the content of the first and second substances,l target is the target name in the target feature vector
The vector of text words of (a),
is the firstnA target name text word vector for each target knowledge base,
it is shown that the calculation of the degree of similarity,ηin order to associate the threshold value with the threshold value,H 0 for the case where the target is not associated,H 1 indicating the case where the target association was successful,N target representing a target quantity of the target knowledge base; that is, the above expression indicates when the minimum value in the similarity calculation results is greater than the correlation threshold valueηIf the target is not associated with the targetH 0 (ii) a When the minimum value in the similarity calculation results is less than the associated threshold valueηIf so, the target association is successfulH 1 (ii) a Based on the result of successful target association, i.e. the above formula result is the case of successful target associationH 1 Automatically extracting indexes related to the target ability and the power range in the knowledge base, and visually presenting the indexes in the situation, so as to complete the association of the situation and the target knowledge base;
finally, the statistical data of the geographic environment, the electromagnetic environment, the natural resources and the humanistic geography in the battlefield environment are visually displayed in a chart mode to form a real-time regional target situation chart;
s302, data information association:
firstly, a related target system module communicates situation with a background knowledge base, a self-adaptive fuzzy related algorithm is adopted to set target country, target name, target model, target membership and space-time attribute as related objects, the targets in the target situation are intelligently related with a comprehensive guarantee system, a command control system, an information support system and a fire striking system of the targets in the background knowledge base in a self-adaptive fuzzy manner, and the system composition distribution, the system strong weakness and the status of the targets in the system are related and presented on a regional target situation map;
secondly, the associated dynamic information module communicates the situation with a background newspaper library, takes a target name, space-time information and a target country as associated objects, extracts associated element items in the dynamic information, intelligently associates the target in the target situation with the dynamic information in the newspaper library according to a self-adaptive fuzzy association algorithm, comprises historical dynamic information and recent dynamic information, presents the information on the situation map in a statistical chart form, and supports trend analysis;
and finally, the associated major event module communicates the situation with a background newspaper library, performs entity extraction on major events by utilizing the ending participle, wherein the entity extraction comprises people, organizations, places, time, targets and national information, and stores the extracted entity objects in major event feature vectors in the following representation form:
E d =[e people ,e country ,e organization ,e place ,e time ,e target ,…]
wherein E is d In order to be a significant event feature vector,e people is a key element of human beings and objects,e country is a key element of the state,e organization is used as the element of the organization mechanism,e place as the location element, the location information is,e time as a time element, the time period of the time period,e target is a target element;
displaying corresponding major events in different regions on a GIS ball according to the spatio-temporal information extracted from the major events and a longitude and latitude comparison table, displaying event development veins in a fishbone graph mode according to the occurrence time of each major event, and classifying and organizing each major event according to different subjects by using a trained TextCNN classifier model to form a subject thematic text data management mode, so that a user can conveniently look up and browse;
meanwhile, a semantic similarity correlation algorithm is adopted to intelligently correlate the target situation with the current global major events in the newspaper library, so that the multidimensional information data are integrally presented, and trend analysis is supported;
s303, generating comprehensive situation information
Based on the steps S301 and S302, the related data information is associated and organized, and the comprehensive situation information is generated through manual evidence judgment, so that the comprehensive situation information not only has the environment and target states in the traditional situation, but also comprises other related important data information related to the target, and support is provided for user analysis and prediction.
In some embodiments, the adaptive fuzzy association algorithm in step S302 is: and taking the target model, the target name, the target membership, the time-space information and the target country element of the target in the situation information as association bodies and taking the association bodies as a sequence, sequentially carrying out similarity calculation based on Euclidean distance on each target, judging that the data information is associated to the target if the similarity obtained by calculation is smaller than an association threshold, and ending the self-adaptive fuzzy association process until a certain association body is associated to all data or all the association bodies are traversed.
In some embodiments, the method for intelligently associating the target situation with the current global significant event in the newspaper library by using the semantic similarity association algorithm in step S302 is as follows:
taking a target name and a target model as an association body, taking each entry in a major event as a similarity calculation object, taking the major event as a sample, and calculating the similarity between the entries by using the complementarity between the sample similarity and the occurrence similarity, wherein the specific formula is as follows:
wherein G is the number of significant event samples,w k is a first of the associated ontologykThe number of the entries is the same as the number of the entries,w j for the second time that similarity needs to be calculated in a significant event samplejThe number of the entries is the same as the number of the entries,T k andT j is composed ofw k Andw j a set of significant event samples of (a) time,x k andy j respectively comprisew k Andw j the significant event text vector features of (1),λa compensation constant of 0 to 1, N: (w k ) And N (w j ) Respectively is composed ofw k Andw j the significant event number of (S) ((S))w k ,w j ) Is composed ofw k Andw j comprehensive similarity of (S) e (w k ,w j ) To comprisew k Andw j similarity between them, S c (w k ,w j ) Is composed ofw k Andw j the co-occurrence similarity distance between them,Nfor text vector featuresx k Andy j dimension of, σ 2 A constant term with a value of 0-1;
② calculation based on comprehensive similarityw k Andw j conditional transition probability of (2):
wherein the content of the first and second substances,p k j, is conditional transition probability, S: (w k ,w j ) As entriesw k Andw j the overall degree of similarity of (a) to (b),Kthe number of entries in the major event;
determining the number of entries with high association degree with the association ontology in the important events needing to be mined based on a random walk methodMIteratively calculating the correlation degree of each entry in the association ontology and the major event until the results of two adjacent iterations are less thanεAnd finishing the iteration process to obtain the relevancy of each entry and sequencing the entries, and taking the topMThe entries serve as the extension entries of the association ontology, and the specific iterative formula is as follows:
wherein the content of the first and second substances,r m (j) Is the first in a major eventjThe first of the entrymThe value of the sub-iteration is,αin order to randomly walk the step size,N I in order to associate the number of ontologies,r m-1 (j) Is the first in a major eventjThe first of the entrym-1 iteration value of the number of iterations,v j indicating the first in a significant eventjInitial probability of each entry relevance degree;
based on the association ontology and the extension entries thereof as the association objects, searching the significant events existing in the association objects, associating the significant events with the targets, and realizing the significant events associated with the targets or the significant events associated with the significant events.
In some embodiments, the significant event is the firstjThe initial probability of the relevance of each entry is set as:
wherein the content of the first and second substances,N I in order to associate the number of ontologies,Min order to mine the number of entries with high association degree with the association ontology in the major event,βis an empirical constant value whenN I When the value is not less than 0, the reaction time is not less than 0,β=0, otherwise
。
In summary, due to the adoption of the technical scheme, the invention has the beneficial effects that:
1. the invention introduces other related data information except the traditional situation data, extracts the entity element characteristics of the related data information by using a natural language processing method, supports the intelligent association of the situation information with a target system, dynamic information and current major events, generates comprehensive situation information with comprehensive contents and multi-dimensional stereo, and enriches the elements and connotation of the traditional situation information.
2. The invention relates the situation and other data information by using similarity calculation, completes the space-time mapping of the major event by extracting space-time element information through entities, realizes the organization and management of the major event according to the subject semantic classification, forms situation comprehensive information, predicts the situation in the situation by using the forms of charts and the like, and compared with the traditional situation information, the invention only has single information of target position, state and the like, and the new situation information relates the target system, dynamic information, major event and the like, expands the information content, and the comprehensive environment superposes the human geography on the basis of the traditional geographic environment and the natural geographic environment, thereby having the characteristics of comprehensiveness, three-dimension, multi-dimension and the like and laying a solid foundation for the subsequent situation analysis and prediction of users.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention, and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
Fig. 1 is a flowchart of an intelligent auxiliary comprehensive situation generating method according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. The components of embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.
Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, belong to the protection scope of the present invention.
Examples
As shown in fig. 1, this embodiment provides an intelligent auxiliary generation method for comprehensive situation, which includes the following steps:
s100, the comprehensive environment construction module presents the geographic environment, the electromagnetic environment, the natural resources and the humanistic geography in a characteristic vector form, and the comprehensive environment is overlapped on the digital earth layer by layer to form a multilayer overlapped comprehensive environment:
(1) geographical environment
Constructing geographic environment feature vectors of various countries according to the geographic environment data, wherein elements of the geographic environment feature vectors comprise distribution features of important facilities such as important ports, important organizations, strategic key roads and the like; thus, the geographic environment feature vector is represented by the form:
wherein the content of the first and second substances,
is as followsiThe country is in the agetThe feature vector of the geographical environment of (a),tin the course of the time of the year,Nis the number of countries;
is as followsiOne countryFamily agetThe important port distribution characteristic vector;
is as followsiThe country is in the agetThe important organizational structure distribution feature vector;
are respectively the firstiThe country is in the agetThe strategic key path distribution feature vector of (1) mainly comprises the contents of the respective geographic position distribution, position action level, facility composition distribution and the like of the important facilities.
(2) Electromagnetic environment
Constructing an electromagnetic environment characteristic vector according to electromagnetic environment data, wherein elements of the electromagnetic environment characteristic vector comprise a channel attenuation coefficient, an electromagnetic interference coefficient, anti-interference capability and spectrum utilization rate; whereby the electromagnetic environment feature vector is represented in the form:
wherein the content of the first and second substances,
is as followspIs in the first areattThe feature vector of the electromagnetic environment at the time of day,
is as followspIs in the first areattThe channel attenuation coefficient characteristics at the time of day,
is as followspIs in the first areattThe characteristic of the electromagnetic interference coefficient at the moment,
is as followspIs in the first areattThe characteristics of the anti-interference capability at the moment,
is as followspIs in the first areattA spectral usage characteristic of the time of day.
(3) Natural resources
Constructing a natural resource feature vector according to the natural resource data, wherein elements of the natural resource feature vector comprise key mineral products and energy; thus, the natural resource feature vector is represented in the form:
wherein, the first and the second end of the pipe are connected with each other,
is as followsiThe country is in the agetThe natural resource feature vector of (2);
、
is as followsiThe country is in the agetThe key mineral distribution characteristic vector and the energy distribution characteristic vector mainly comprise the contents of reserves, geographical positions and the like of resources such as iron ores, rare earth, petroleum, coal and the like.
(4) Humanistic geography
Constructing a human geographical feature vector according to human geographical data, wherein elements of the human geographical feature vector comprise population number, religion number and average education year number; thus, the human geographic feature vector is represented in the form of:
wherein the content of the first and second substances,tin the course of the time of the year,
is as followsiThe country is in the agetThe human-geographic feature vector of (a),Nthe number of countries is the number of countries,
is as followsiThe country is in the agetThe number of population of (a) is,
is as followsiThe country is in the agetThe elements of the religious quantity of (c),
is as followsiThe country is in the yeartAverage educational years element of [ … ]] T Is the transposition of the vector;
combining the humanistic geographic feature vectors of all countries to form a world humanistic geographic feature matrix, wherein the expression form is as follows:
wherein the content of the first and second substances,tin the order of the years, the number of the years,
is a world humanistic geographic feature matrix,
is as followsiThe country is in the agetThe human geographic feature vector of (1); elements in the world humanity geographic feature matrix are stored in a database table form.
S200, the target situation construction module generates target situation feature vectors from the real-time target data of the region:
the dynamic elements of the target situation feature vector comprise a target type, longitude, latitude, altitude and speed; therefore, the target situation feature vector is expressed in the form of:
wherein the content of the first and second substances,
is as followsiiThe object is atttTarget situation feature vectors of moments;
denotes the firstiiThe object is atttA target type of time;
is shown asiiThe object is atttA target longitude of the time of day;
is shown asiiThe object is atttTarget dimensions of time;
is shown asiiThe object is atttA target height at a time;
is shown asiiThe object is atttTarget speed at the moment.
S300, the data association and synthesis module depends on a multi-layer superposed synthetic environment and combines target situation characteristic vectors, intelligent generation of target situations is achieved through the modes of entity extraction, topic association, space-time mapping and semantic association, regional target situation is associated to dynamic information, a target system and major events, and finally comprehensive situation information of a full-dimensional multi-domain is formed:
s301, generating a target situation:
firstly, a target space state is presented on the digital earth in a target eigenvector form on the basis of the digital earth and a multi-layer superposed comprehensive environment, elements of the target eigenvector comprise space-time attributes, target countries, target names, target models and target membership, and the target eigenvector is represented in the form:
wherein the content of the first and second substances,
is as followsiiThe object is atttTarget feature vectors of the time;
is as followsiiThe object is atttThe target situation characteristic vector of the moment comprises space-time attributes and target model elements;
is used as the target membership element and is provided with a plurality of target membership elements,
as the elements of the name of the object,
is a target country element;
secondly, the situation is associated with a target knowledge base, similarity calculation is carried out according to the target name in the target characteristic vector and the target name in the knowledge base, information of the relevant target knowledge base is associated based on a similarity calculation result, and a calculation formula is as follows:
wherein the content of the first and second substances,l target is the target name in the target feature vector
The vector of text words of (a),
is the firstnA target name text word vector for each target knowledge base,
the calculation of the degree of similarity is indicated,ηin order to associate the threshold value with the threshold value,H 0 for the case where the target is not associated,H 1 indicating the case where the target association was successful,N target representing a target quantity of the target knowledge base; that is, the above expression indicates when the minimum value in the similarity calculation results is greater than the correlation threshold valueηIf the target is not associated with the targetH 0 (ii) a When the minimum value in the similarity calculation results is less than the associated threshold valueηIf so, the target association is successfulH 1 (ii) a Based on the result of successful target association, i.e. the above formula result is the case of successful target associationH 1 Automatically extracting indexes related to the target ability and the power range in the knowledge base, and visually presenting the indexes in the situation, so as to complete the association of the situation and the target knowledge base;
finally, the statistical data of the geographic environment, the electromagnetic environment, the natural resources and the humanistic geography in the battlefield environment are visually displayed in a chart mode to form a real-time regional target situation chart;
s302, data information association:
firstly, a related target system module communicates situation with a background knowledge base, a self-adaptive fuzzy related algorithm is adopted to set target country, target name, target model, target membership and space-time attribute as related objects, the targets in the target situation and a comprehensive guarantee system (composition distribution, strong weakness and status action) of the targets in the background knowledge base are intelligently related in a self-adaptive fuzzy manner, a command control system, an information support system and a fire striking system are related, and the composition distribution, the strong weakness and the status action of the targets in the system are related and presented on a regional target situation map; the adaptive fuzzy association algorithm is as follows: and taking the target model, the target name, the target membership, the time-space information and the target country element of the target in the situation information as association bodies and taking the association bodies as a sequence, sequentially carrying out similarity calculation based on Euclidean distance on each target, judging that the data information is associated to the target if the similarity obtained by calculation is smaller than an association threshold until a certain association body is associated to all data or all the association bodies are traversed, and finishing the self-adaptive fuzzy association process.
In this embodiment, the implementation of the adaptive fuzzy association algorithm is as follows:
the associated objects of the self-adaptive fuzzy association algorithm comprise target countries, target names, target models, target membership and space-time attributes;
the input of the adaptive fuzzy association algorithm comprises
、
、
、
、
、
、
、
、
、
、
、
、
、
And
；
the flow of the adaptive fuzzy association algorithm is as follows:
the output of the adaptive fuzzy correlation algorithm is
；
Wherein the content of the first and second substances,
is as followsiiThe object is atttTarget characteristic vectors of a related target system in time;Ris a target number;
、
、
、
、
respectively representiiTarget model, target name, target membership, space-time information and text word vector of target country of each target;
、
、
、
and
respectively representtThe first of a target systemiThe target type, the target name, the target membership, the time-space information and the text word vector of the target country; e t_s Target system feature vectors on all the associations are obtained;
is as followsiiThe object is atttTarget feature vectors within a moment;N t_tixi is the number of target systems;
、
、
、
and
respectively is a target model, a target name, target membership, space-time information and a target country similarity correlation threshold.
Secondly, the associated dynamic information module communicates the situation with a background newspaper library, takes a target name, space-time information and a target country as associated objects, extracts associated element items in the dynamic information, intelligently associates the target in the target situation with the dynamic information in the newspaper library according to a self-adaptive fuzzy association algorithm, comprises historical dynamic information and recent dynamic information, presents the information on the situation map in a statistical chart form, and supports trend analysis;
and finally, the associated major event module communicates the situation with a background newspaper library, performs entity extraction on major events by utilizing the ending participle, wherein the entity extraction comprises people, organizations, places, time, targets and national information, and stores the extracted entity objects in major event feature vectors in the following representation form:
E d =[e people ,e country ,e organization ,e place ,e time ,e target ,…]
wherein E is d In order to be a significant event feature vector,e people is an element of a human body and is a human body,e country is a key element of the state,e organization is used as the element of the organization mechanism,e place as the location element, the location information is,e time as a time element, the time period of the time period,e target is a target element;
displaying corresponding major events in different regions on a GIS ball according to the spatio-temporal information extracted from the major events and a longitude and latitude comparison table, displaying event development veins in a fishbone graph mode according to the occurrence time of each major event, and classifying and organizing each major event according to different subjects by using a trained TextCNN classifier model to form a subject thematic text data management mode, so that a user can conveniently look up and browse;
meanwhile, a semantic similarity correlation algorithm is adopted to intelligently correlate the target situation with the current global major events in the newspaper library, so that the multidimensional information data are integrally presented, and trend analysis is supported; specifically, the method comprises the following steps:
taking a target name and a target model as an association body, taking each entry in a major event as a similarity calculation object, taking the major event as a sample, and calculating the similarity between the entries by using the complementarity between the sample similarity and the occurrence similarity, wherein the specific formula is as follows:
wherein G is the number of significant event samples,w k is a first of the associated ontologykThe number of the entries is the same as the number of the entries,w j for the second time that similarity needs to be calculated in a significant event samplejThe number of the entries is the same as the number of the entries,T k andT j is composed ofw k Andw j a set of significant event samples of (a) time,x k andy j respectively is composed ofw k Andw j the significant event text vector features of (1),λa compensation constant of 0 to 1, N: (w k ) And N (w j ) Respectively is composed ofw k Andw j of significant events, S: (w k ,w j ) Is composed ofw k Andw j comprehensive similarity of (S) e (w k ,w j ) To comprisew k Andw j similarity between them, S c (w k ,w j ) Is composed ofw k Andw j the co-occurrence similarity distance between them,Nfor text vector featuresx k Andy j dimension of, σ 2 And constant terms with values of 0-1.
② calculation based on comprehensive similarityw k Andw j conditional transition probability of (2):
wherein the content of the first and second substances,p k j, is a conditional transition probability, S: (w k ,w j ) As entriesw k Andw j the overall degree of similarity of (a) to (b),Kthe number of entries in the major event;
thirdly, based on a random walk method,determining the number of entries with high association degree with an association ontology in important events needing to be minedMIteratively calculating the correlation degree of each entry in the association ontology and the major event until the results of two adjacent iterations are less thanεAnd finishing the iteration process to obtain the relevancy of each entry and sequencing the entries, and taking the topMThe entries are used as the extension entries of the association body, the iteration process is finished, the relevancy of each entry is obtained and sequenced, and the specific iteration formula is as follows:
wherein the content of the first and second substances,r m (j) Is the first in a major eventjThe first of the entrymThe value of the sub-iteration is,αin order to randomly walk the step size,N I in order to associate the number of ontologies,r m-1 (j) Is the first in a major eventjThe first of the entrym-1 iteration value of the number of iterations,v j indicating the first in a significant eventjInitial probability of each entry relevance degree; the general settings are:
wherein the content of the first and second substances,N I in order to associate the number of ontologies,Min order to mine the number of entries with high association degree with the association ontology in the major event,βis an empirical constant value whenN I When the value is not less than 0, the reaction time is not less than 0,β=0, otherwise
。
And fourthly, searching the significant events existing in the associated objects based on the associated ontology and the extension entries thereof as the associated objects, and associating the significant events with the targets to realize the significant events associated with the targets or the significant events associated with the targets.
S303, generating comprehensive situation information
Based on the steps S301 and S302, the related data information is associated and organized, and the comprehensive situation information is generated through manual evidence judgment, so that the comprehensive situation information not only has the environment and the target state in the traditional situation, but also comprises other related important data information related to the target, and support is provided for user analysis and prediction.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. An intelligent auxiliary comprehensive situation generation method is characterized by comprising the following steps:
s100, a comprehensive environment construction module presents the geographic environment, the electromagnetic environment, the natural resources and the humanistic geography in a characteristic vector form and superposes the geographic environment, the electromagnetic environment, the natural resources and the humanistic geography on the digital earth layer by layer to form a multilayer superposed comprehensive environment;
s200, a target situation construction module generates target situation characteristic vectors from the real-time target data of the region;
s300, the data association and synthesis module depends on a multi-layer overlapped synthetic environment and combines target situation characteristic vectors to realize intelligent generation of target situations and association of regional target situations to information, a target system and major events through the ways of entity extraction, topic association, space-time mapping and semantic association, and finally comprehensive situation information of full-dimensional multi-domain is formed.
2. The intelligent auxiliary comprehensive situation generation method according to claim 1, wherein the method for presenting the geographic environment in the form of feature vectors in step S100 is as follows:
constructing geographic environment feature vectors of various countries according to the geographic environment data, wherein the elements of the geographic environment feature vectors comprise important ports, important organizations and strategic important roads; thus, the geographic environment feature vector is represented by the form:
wherein the content of the first and second substances,
is as followsiThe country is in the agetThe feature vector of the geographical environment of (a),tin the course of the time of the year,Nis the number of countries;
is as followsiThe country is in the agetThe important port distribution characteristic vector;
is as followsiThe country is in the agetThe important organizational structure distribution feature vector;
are respectively the firstiThe country is in the agetThe strategic key path of (1) distributes feature vectors.
3. The intelligent auxiliary comprehensive situation generation method according to claim 1, wherein the method for presenting the electromagnetic environment in the form of feature vectors in step S100 is as follows:
constructing an electromagnetic environment characteristic vector according to electromagnetic environment data, wherein elements of the electromagnetic environment characteristic vector comprise a channel attenuation coefficient, an electromagnetic interference coefficient, anti-interference capability and spectrum utilization rate; whereby the electromagnetic environment feature vector representation is of the form:
wherein, the first and the second end of the pipe are connected with each other,
is as followspIs in the first areattThe feature vector of the electromagnetic environment at the time of day,
is as followspIs in the first areattThe channel attenuation coefficient characteristics at the time of day,
is as followspIs in the first areattThe characteristic of the electromagnetic interference coefficient at a time,
is as followspIs in the first areattThe characteristics of the anti-interference capability at the moment,
is as followspIs in the first areattSpectral usage characteristics at a time.
4. The intelligent auxiliary generation method for comprehensive situations according to claim 1, wherein the method for presenting natural resources in a feature vector form in step S100 is:
constructing a natural resource feature vector according to the natural resource data, wherein elements of the natural resource feature vector comprise key mineral products and energy; thus, the natural resource feature vector is represented in the form:
wherein the content of the first and second substances,
is a firstiThe country is in the agetThe natural resource feature vector of (2);
、
is as followsiThe country is in the yeartThe key mineral distribution eigenvectors and the energy distribution eigenvectors.
5. The intelligent auxiliary comprehensive situation generation method according to claim 1, wherein the method for presenting the human geography in the form of feature vectors in step S100 is as follows:
constructing a human geographical feature vector according to human geographical data, wherein elements of the human geographical feature vector comprise population number, religion number and average education year number; thus, the human geographic feature vector is represented in the form of:
wherein the content of the first and second substances,tin the course of the time of the year,
is a firstiThe country is in the agetThe human-like geographic feature vector of (a),Nthe number of countries is the number of countries,
is as followsiThe country is in the agetThe number of population of (a) is,
is as followsiThe country is in the agetThe elements of the religious quantity of (c),
is a firstiThe country is in the yeartAverage educational age factor, [ … ]] T Is the transposition of the vector;
combining the humanistic geographic feature vectors of all countries to form a world humanistic geographic feature matrix, wherein the expression form is as follows:
wherein the content of the first and second substances,tin the course of the time of the year,
is a world human geographic feature matrix,
is as followsiThe country is in the agetThe human geographic feature vector of (1); elements in the world humanity geographic feature matrix are stored in a database table form.
6. The intelligent auxiliary comprehensive situation generation method according to claim 1, wherein the method for the target situation construction module to generate the target situation feature vector from the regional real-time target data in step S200 is as follows:
the dynamic elements of the target situation feature vector comprise a target type, longitude, latitude, altitude and speed; therefore, the target situation feature vector is expressed in the form of:
wherein the content of the first and second substances,
is as followsiiThe object is atttTarget situation feature vectors of moments;
is shown asiiThe object is atttA target type of time;
is shown asiiThe object is atttA target longitude of the time of day;
is shown asiiThe object is atttTarget dimensions of time;
is shown asiiThe object is atttA target height at a time;
is shown asiiThe object is atttThe target speed at the moment.
7. The intelligent auxiliary comprehensive situation generation method according to claim 1, wherein step S300 comprises the following substeps:
s301, generating a target situation:
firstly, a target space state is presented on the digital earth in a target eigenvector form on the basis of the digital earth and a multi-layer superposed comprehensive environment, elements of the target eigenvector comprise space-time attributes, target countries, target names, target models and target membership, and the target eigenvector is represented in the form:
wherein the content of the first and second substances,
is as followsiiThe object is atttTarget feature vectors at a moment;
is as followsiiThe object is atttThe target situation characteristic vector of the moment comprises space-time attributes and target model elements;
is used as the target membership element and is provided with a plurality of target membership elements,
as the elements of the name of the object,
is a target country element;
secondly, the situation is associated with a target knowledge base, similarity calculation is carried out according to the target name in the target characteristic vector and the target name in the knowledge base, information of the relevant target knowledge base is associated based on the similarity calculation result, and the calculation formula is as follows:
wherein the content of the first and second substances,l target is the target name in the target feature vector
The vector of text words of (a),
is the firstnA target name text word vector for each target knowledge base,
it is shown that the calculation of the degree of similarity,ηin order to associate the threshold value with the threshold value,H 0 for the case where the target is not associated,H 1 indicating the case where the target association was successful,N target representing a target quantity of the target knowledge base; that is, the above expression indicates when the minimum value in the similarity calculation results is greater than the correlation threshold valueηIf the target is not associated with the targetH 0 (ii) a When the minimum value in the similarity calculation results is less than the associated threshold valueηIf the target association is successful, the target association is performedH 1 (ii) a Based on the result of successful target association, i.e. the above formula result is the case of successful target associationH 1 Automatically extracting indexes related to the target ability and the power range in the knowledge base, and visually presenting the indexes in the situation, so as to complete the association of the situation and the target knowledge base;
finally, the statistical data of the geographic environment, the electromagnetic environment, the natural resources and the humanistic geography in the battlefield environment are visually displayed in a chart mode to form a real-time regional target situation chart;
s302, data information association:
firstly, a related target system module communicates situation with a background knowledge base, a self-adaptive fuzzy related algorithm is adopted to set target country, target name, target model, target membership and space-time attribute as related objects, the targets in the target situation are intelligently related with a comprehensive guarantee system, a command control system, an information support system and a fire striking system of the targets in the background knowledge base in a self-adaptive fuzzy manner, and the system composition distribution, the system strong weakness and the status of the targets in the system are related and presented on a regional target situation map;
secondly, the associated dynamic information module communicates the situation with a background newspaper library, takes a target name, space-time information and a target country as associated objects, extracts associated element items in the dynamic information, intelligently associates the target in the target situation with the dynamic information in the newspaper library according to a self-adaptive fuzzy association algorithm, comprises historical dynamic information and recent dynamic information, presents the information on the situation map in a statistical chart form, and supports trend analysis;
and finally, the associated major event module communicates the situation with a background newspaper library, performs entity extraction on major events by utilizing the ending participle, wherein the entity extraction comprises people, organizations, places, time, targets and national information, and stores the extracted entity objects in major event feature vectors in the following representation form:
E d =[e people ,e country ,e organization ,e place ,e time ,e target ,…]
wherein E is d Is heavyThe feature vector of the large event is used,e people is an element of a human body and is a human body,e country is a key element of the state,e organization is used as the element of the organization mechanism,e place as the location element, the location information is,e time as a time element, the time period of the time period,e target is a target element;
displaying corresponding major events in different regions on a GIS ball according to the spatio-temporal information extracted from the major events and a longitude and latitude comparison table, displaying event development veins in a fishbone graph mode according to the occurrence time of each major event, and classifying and organizing each major event according to different subjects by using a trained TextCNN classifier model to form a subject thematic text data management mode, so that a user can conveniently look up and browse;
meanwhile, a semantic similarity correlation algorithm is adopted to intelligently correlate the target situation with the current global major events in the newspaper library, so that the multidimensional information data are integrally presented, and trend analysis is supported;
s303, generating comprehensive situation information
Based on the steps S301 and S302, the related data information is associated and organized, and the comprehensive situation information is generated through manual evidence judgment, so that the comprehensive situation information not only has the environment and the target state in the traditional situation, but also comprises other related important data information related to the target, and support is provided for user analysis and prediction.
8. The intelligent assisted comprehensive situation generation method according to claim 7, wherein the adaptive fuzzy association algorithm in step S302 is: and taking the target model, the target name, the target membership, the time-space information and the target country element of the target in the situation information as association bodies and taking the association bodies as a sequence, sequentially carrying out similarity calculation based on Euclidean distance on each target, judging that the data information is associated to the target if the similarity obtained by calculation is smaller than an association threshold, and ending the self-adaptive fuzzy association process until a certain association body is associated to all data or all the association bodies are traversed.
9. The intelligent auxiliary comprehensive situation generation method according to claim 7, wherein the method for intelligently associating the target situation with the current global significant event in the newspaper library by using the semantic similarity association algorithm in step S302 is as follows:
taking a target name and a target model as an association body, taking each entry in a major event as a similarity calculation object, taking the major event as a sample, and calculating the similarity between the entries by using the complementarity between the sample similarity and the occurrence similarity, wherein the specific formula is as follows:
wherein G is the number of significant event samples,w k is a first of the associated ontologykThe number of the entries is the same as the number of the entries,w j for the second time that similarity needs to be calculated in a significant event samplejThe number of the entries is the same as the number of the entries,T k andT j is composed ofw k Andw j a set of significant event samples of (a) time,x k andy j respectively is composed ofw k Andw j the significant event text vector features of (1),λa compensation constant of 0 to 1, N: (w k ) And N (w j ) Respectively is composed ofw k Andw j the significant event number of (S) ((S))w k ,w j ) Is composed ofw k Andw j comprehensive similarity of (S) e (w k ,w j ) To comprisew k Andw j similarity between them, S c (w k ,w j ) Is composed ofw k Andw j the co-occurrence similarity distance between them,Nfor text vector featuresx k Andy j dimension of, σ 2 A constant term with a value of 0-1;
② calculation based on comprehensive similarityw k Andw j conditional transition probability of (2):
wherein the content of the first and second substances,p k j, is a conditional transition probability, S: (w k ,w j ) As entriesw k Andw j the overall degree of similarity of (a) to (b),Kthe number of entries in the major event;
determining the number of entries with high association degree with the association ontology in the important events needing to be mined based on a random walk methodMIteratively calculating the correlation degree of each entry in the association ontology and the major event until the results of two adjacent iterations are less thanεAnd finishing the iteration process to obtain the relevancy of each entry and sequencing the entries, and taking the topMThe entries serve as the extension entries of the association ontology, and the specific iterative formula is as follows:
wherein the content of the first and second substances,r m (j) Is the first in a major eventjThe first of the entrymThe value of the sub-iteration is,αin order to randomly walk the step size,N I in order to associate the number of ontologies,r m-1 (j) Is the first in a major eventjThe first of the entrym-1 iteration value of the number of iterations,v j indicating the first in a significant eventjInitial probability of each entry relevance degree;
based on the association ontology and the extension entries thereof as the association objects, searching the significant events existing in the association objects, associating the significant events with the targets, and realizing the significant events associated with the targets or the significant events associated with the significant events.
10. The intelligent assisted-generation method for comprehensive situation according to claim 9, characterized in that the second in the significant eventjThe initial probability of the relevance of each entry is set as:
wherein the content of the first and second substances,N I in order to associate the number of ontologies,Min order to mine the number of entries with high association degree with the association ontology in the major event,βis an empirical constant value whenN I When the pressure is not higher than 0, the pressure is lower than 0,β=0, otherwise
。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210914211.0A CN114969350B (en) | 2022-08-01 | 2022-08-01 | Intelligent auxiliary generation method for comprehensive situation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210914211.0A CN114969350B (en) | 2022-08-01 | 2022-08-01 | Intelligent auxiliary generation method for comprehensive situation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114969350A true CN114969350A (en) | 2022-08-30 |
| CN114969350B CN114969350B (en) | 2022-11-22 |
Family
ID=82969412
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210914211.0A Active CN114969350B (en) | 2022-08-01 | 2022-08-01 | Intelligent auxiliary generation method for comprehensive situation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114969350B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102035678A (en) * | 2010-12-16 | 2011-04-27 | 中国电子科技集团公司第三十研究所 | Multidimensional comprehensive situation display system based on degree of association |
| CN104680260A (en) * | 2015-03-17 | 2015-06-03 | 云南电力调度控制中心 | Method and system for establishing power grid visualized perceptual model |
| US20160021141A1 (en) * | 2014-07-18 | 2016-01-21 | The Regents Of The University Of Michigan | Rating network security posture and comparing network maliciousness |
| CN107239562A (en) * | 2017-06-13 | 2017-10-10 | 华东理工大学 | The analysis of public opinion method associated based on probability characteristics |
| CN111460730A (en) * | 2020-03-26 | 2020-07-28 | 中国电子科技集团公司第二十八研究所 | Future combat intelligent technology application design method |
| CN112966912A (en) * | 2021-02-25 | 2021-06-15 | 李昌玺 | Real-time battlefield situation updating algorithm |
| CN113255916A (en) * | 2021-05-31 | 2021-08-13 | 成都卡莱博尔信息技术股份有限公司 | Using method of combat knowledge system and combat decision auxiliary method |
| CN114139499A (en) * | 2022-02-08 | 2022-03-04 | 中国电子科技集团公司第五十四研究所 | Situation text message semantic compression coding method based on TSS coding |
| CN114139550A (en) * | 2022-02-08 | 2022-03-04 | 中国电子科技集团公司第五十四研究所 | Situation intelligent cognition method based on activity semantic text message |
-
2022
- 2022-08-01 CN CN202210914211.0A patent/CN114969350B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102035678A (en) * | 2010-12-16 | 2011-04-27 | 中国电子科技集团公司第三十研究所 | Multidimensional comprehensive situation display system based on degree of association |
| US20160021141A1 (en) * | 2014-07-18 | 2016-01-21 | The Regents Of The University Of Michigan | Rating network security posture and comparing network maliciousness |
| CN104680260A (en) * | 2015-03-17 | 2015-06-03 | 云南电力调度控制中心 | Method and system for establishing power grid visualized perceptual model |
| CN107239562A (en) * | 2017-06-13 | 2017-10-10 | 华东理工大学 | The analysis of public opinion method associated based on probability characteristics |
| CN111460730A (en) * | 2020-03-26 | 2020-07-28 | 中国电子科技集团公司第二十八研究所 | Future combat intelligent technology application design method |
| CN112966912A (en) * | 2021-02-25 | 2021-06-15 | 李昌玺 | Real-time battlefield situation updating algorithm |
| CN113255916A (en) * | 2021-05-31 | 2021-08-13 | 成都卡莱博尔信息技术股份有限公司 | Using method of combat knowledge system and combat decision auxiliary method |
| CN114139499A (en) * | 2022-02-08 | 2022-03-04 | 中国电子科技集团公司第五十四研究所 | Situation text message semantic compression coding method based on TSS coding |
| CN114139550A (en) * | 2022-02-08 | 2022-03-04 | 中国电子科技集团公司第五十四研究所 | Situation intelligent cognition method based on activity semantic text message |
Non-Patent Citations (10)
| Title |
|---|
| 上海勇进软件: ""航天战场态势感知平台的功能介绍"", 《HTTP://WWW.UIPOWER.COM/NEWS/BATTLEFIELD.HTML》 * |
| 孙洪星: ""飞行仿真训练系统综合控制台三维态势显示的设计与实现"", 《中国优秀硕士学位论文全文数据库 社会科学Ⅰ辑》 * |
| 朱丰等: "基于深度学习的战场态势评估综述与研究展望", 《军事运筹与系统工程》 * |
| 李均阁: "多源信息融合及其应用", 《甘肃科技》 * |
| 杨宝强等: "数据融合中态势估计的模型建立与实现", 《空军工程大学学报(自然科学版)》 * |
| 汪跃等: "战场态势一张图技术综述", 《指挥信息系统与技术》 * |
| 王瑞霞: "频谱监测系统电磁态势感知模块设计", 《电子质量》 * |
| 臧勤 等: ""一种基于数据库的综合态势生成方法"", 《雷达与对抗》 * |
| 薛本新: "通用战场环境态势图研究", 《军事运筹与系统工程》 * |
| 陆强强等: "一种利用目标属性进行态势关联的方法", 《火力与指挥控制》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114969350B (en) | 2022-11-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP7468929B2 (en) | How to acquire geographical knowledge | |
| CN104318340B (en) | Information visualization methods and intelligent visible analysis system based on text resume information | |
| CN103390039B (en) | Urban disaster thematic map real-time generating method based on network information | |
| CN106095762A (en) | A kind of news based on ontology model storehouse recommends method and device | |
| CN111382276B (en) | Event development context graph generation method | |
| CN107423820B (en) | Knowledge graph representation learning method combined with entity hierarchy categories | |
| Lee | Unsupervised and supervised learning to evaluate event relatedness based on content mining from social-media streams | |
| Wang et al. | Preference-based spatial co-location pattern mining | |
| CN112328794B (en) | Typhoon event information aggregation method | |
| CN112559907A (en) | Basic data retrieval and integrated display method based on spatio-temporal label spatio-temporal correlation | |
| CN103455487A (en) | Extracting method and device for search term | |
| CN113672599A (en) | Visual aid decision-making method for realizing government affair informatization project construction management by creating domain knowledge graph | |
| CN113190593A (en) | Search recommendation method based on digital human knowledge graph | |
| Fasel et al. | A data warehouse model for integrating fuzzy concepts in meta table structures | |
| CN107704524A (en) | A kind of subway station function method for digging based on doc2vec | |
| Wu et al. | Ontology-enriched multi-document summarization in disaster management using submodular function | |
| Lu et al. | Construction of multimedia assisted legal classroom teaching model based on data mining algorithm | |
| Royuela et al. | HouSI: Heuristic for delimitation of housing submarkets and price homogeneous areas | |
| Mair et al. | The grand old party–a party of values? | |
| CN114969350B (en) | Intelligent auxiliary generation method for comprehensive situation | |
| Deraman et al. | A social media mining using topic modeling and sentiment analysis on tourism in Malaysia during COVID19 | |
| CN110909124B (en) | Hybrid enhanced intelligent demand accurate sensing method and system based on human-in-loop | |
| CN102629278B (en) | Semantic annotation and searching method based on problem body | |
| CN109710712B (en) | Case element analysis-based crime hotspot feature mining method and system | |
| Tang et al. | Tourism domain ontology construction from the unstructured text documents |
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 |