CN112241459A - Task-based weapon equipment knowledge graph query and recommendation method and system - Google Patents
Task-based weapon equipment knowledge graph query and recommendation method and system Download PDFInfo
- Publication number
- CN112241459A CN112241459A CN202011136229.XA CN202011136229A CN112241459A CN 112241459 A CN112241459 A CN 112241459A CN 202011136229 A CN202011136229 A CN 202011136229A CN 112241459 A CN112241459 A CN 112241459A
- Authority
- CN
- China
- Prior art keywords
- equipment
- weaponry
- weapon
- label
- task
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 60
- 238000004458 analytical method Methods 0.000 claims abstract description 33
- 238000004140 cleaning Methods 0.000 claims abstract description 12
- 238000001914 filtration Methods 0.000 claims abstract description 11
- 238000005065 mining Methods 0.000 claims abstract description 5
- 238000012800 visualization Methods 0.000 claims abstract description 5
- 238000012423 maintenance Methods 0.000 claims description 28
- 230000006399 behavior Effects 0.000 claims description 26
- 230000001149 cognitive effect Effects 0.000 claims description 23
- 238000010276 construction Methods 0.000 claims description 23
- 230000008569 process Effects 0.000 claims description 22
- 238000007405 data analysis Methods 0.000 claims description 13
- 238000003860 storage Methods 0.000 claims description 12
- 230000008859 change Effects 0.000 claims description 11
- 238000012549 training Methods 0.000 claims description 9
- 238000004422 calculation algorithm Methods 0.000 claims description 8
- 230000003068 static effect Effects 0.000 claims description 8
- 238000011161 development Methods 0.000 claims description 6
- 238000012417 linear regression Methods 0.000 claims description 6
- 238000007621 cluster analysis Methods 0.000 claims description 5
- 230000002159 abnormal effect Effects 0.000 claims description 4
- 230000000694 effects Effects 0.000 claims description 4
- 238000012847 principal component analysis method Methods 0.000 claims description 4
- 230000009471 action Effects 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000010219 correlation analysis Methods 0.000 claims description 3
- 230000003203 everyday effect Effects 0.000 claims description 3
- 230000001131 transforming effect Effects 0.000 claims description 3
- 230000001960 triggered effect Effects 0.000 claims description 3
- 238000005406 washing Methods 0.000 claims 1
- 238000005516 engineering process Methods 0.000 abstract description 7
- 238000007418 data mining Methods 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 7
- 238000010801 machine learning Methods 0.000 description 6
- 238000012545 processing Methods 0.000 description 6
- 230000006872 improvement Effects 0.000 description 4
- 238000007726 management method Methods 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 230000009466 transformation Effects 0.000 description 4
- 206010000117 Abnormal behaviour Diseases 0.000 description 3
- 230000019771 cognition Effects 0.000 description 3
- 230000007423 decrease Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 230000002349 favourable effect Effects 0.000 description 2
- 230000035772 mutation Effects 0.000 description 2
- 230000000750 progressive effect Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000012827 research and development Methods 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 230000002354 daily effect Effects 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
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
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2462—Approximate or statistical queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/25—Integrating or interfacing systems involving database management systems
- G06F16/254—Extract, transform and load [ETL] procedures, e.g. ETL data flows in data warehouses
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/283—Multi-dimensional databases or data warehouses, e.g. MOLAP or ROLAP
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
- G06F16/285—Clustering or classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
Abstract
The invention discloses a task-based weapon equipment knowledge graph query and recommendation method and system, which belong to the field of big data mining, and the technical problem to be solved is how to utilize advanced technologies of internet industries such as knowledge graphs and the like to realize the query and recommendation of the weapon equipment knowledge graph, and the technical scheme is as follows: the weapon equipment ternary entity is used for integrally recognizing the weapon, and performing all-around information display, analysis and mining on the weapon equipment through data standardization and visualization on the basis of gathering basic information and dynamic capability information of the weapon equipment to construct a knowledge graph of the weapon entity; the method comprises the following specific steps: s1, constructing a weapon equipment entity label: extracting basic information of the weaponry entities into basic information tags of the weaponry entities through analysis, filtering and cleaning of the basic information of the weaponry entities; s2, constructing a weapon equipment capability label; s3, constructing a knowledge graph label of the weapon equipment; and S4, constructing a task-based recommended equipment entity label.
Description
Technical Field
The invention relates to the field of big data mining, in particular to a task-based method and a task-based system for inquiring and recommending a weapon equipment knowledge graph.
Background
For many years, weaponry systems have explored the use of information technologies such as knowledge maps to achieve fine management of weaponry and to modify equipment use management during the mission. At present, the main fields of weapon equipment information are basically concentrated in the fields of basic information databases of weapon equipment, current capability query systems of weapon equipment, task statistical analysis systems and the like, but the deep mining, cognition and recommendation work aiming at task types and equipment use is rarely broken through. Therefore, how to utilize advanced technologies of the internet industry such as the knowledge graph and the like to realize inquiry and recommendation of the knowledge graph of the weapon equipment is a problem to be solved urgently at present.
Disclosure of Invention
The technical task of the invention is to provide a task-based weapon equipment knowledge graph query and recommendation method and system, so as to solve the problem of how to utilize advanced technologies of internet industries such as knowledge graphs and the like to realize the query and recommendation of weapon equipment knowledge graphs.
The technical task of the invention is realized in the following way, and the task-based method for inquiring and recommending the knowledge graph of the weapon equipment is characterized in that the weapon equipment ternary entity is used for integrally recognizing the weapon, and on the basis of gathering basic information and dynamic capability information of the weapon equipment, the weapon equipment is subjected to all-round information display, analysis and mining through data standardization and visualization to construct the knowledge graph of the weapon entity; the method comprises the following specific steps:
s1, constructing a weapon equipment entity label: extracting basic information of the weaponry entities into basic information tags of the weaponry entities through analysis, filtering and cleaning of the basic information of the weaponry entities;
s2, constructing a weaponry capability label: extracting the recent dynamic behaviors of the capabilities of the weaponry into capability behavior labels of the weaponry through analysis, filtering and cleaning of the recent dynamic behaviors of the capabilities of the weaponry;
s3, constructing a weapon equipment knowledge graph label: after carrying out comprehensive analysis on a basic information label and an ability behavior label of a weapon equipment entity for data modeling, giving a cognitive label of the equipment ability;
s4, constructing a task recommendation equipment-based entity label: on the basis of the cognitive label of the equipment capability, the equipment knowledge graph construction of the task process is added, and then the equipment with the label is recommended to the completion target of the task, so that the equipment recommendation suggestion is provided for the expansion of the task in the later period.
Preferably, the basic information label is obtained by cleaning and processing heterogeneous data through an ETL tool, so that all data are subjected to two-stage specification according to the display dimension and the analysis dimension, and a data warehouse and an equipment knowledge graph with unified specification are created.
Preferably, the weapon equipment capability label is used for carrying out big data analysis and deep description on the dynamic records of weapon equipment at different periods;
after the weaponry is arranged in the army, receiving a series of activities of using, transforming, maintaining and returning to a factory, generating new information every day, and updating the new information so as to better follow the capacity state change process of the weaponry;
when one weapon equipment is not used for a long time or is frequently used in recent actions, setting thresholds for information of different equipment stages; when the threshold value is triggered, warning is given to the change of the fighting value of the equipment, and a weapon user or a maintainer is reminded to pay attention to the weapon equipment;
preferably, the weapon equipment knowledge graph label is used for respectively setting different influence factors on the static state and the dynamic state of the weapon equipment, repeatedly training and modifying by using big data, selecting six dimensional indexes of a battle scene, the boring attack, the service life, the operation difficulty, the task difficulty and the maintenance difficulty from a plurality of acquisition items, and specifically and measurably analyzing the weapon equipment; and then establishing a six-dimensional attention model of the weaponry by using the six-dimensional indexes, dividing the six-dimensional indexes of the combat scene, the attack tediousness, the service life, the operation difficulty, the task difficulty and the maintenance difficulty into 10 grades, and quantizing the indexes from 1 grade to 10 grades, wherein the larger the index value is, the more important attention is represented by the behavior of the weaponry in the dimension.
Preferably, the establishment of the weapon equipment six-dimensional attention model is as follows:
(1) reducing the n-dimensional model into a 24-dimensional attention model by adopting a principal component analysis method in cluster analysis;
(2) and selecting 6 acquisition items with highest correlation from the acquisition items and the attention models of the weaponry to construct a six-dimensional attention model of the weaponry.
Preferably, the method for analyzing the principal components in the cluster analysis is adopted to reduce the n-dimensional model to a 24-dimensional attention model as follows:
firstly, a K mean algorithm is adopted, the convergence condition is that the error square sum SSE is minimum, and the generated clusters are compact and independent by utilizing the criterion; the equation for the sum of squared errors SEE is as follows:
wherein k represents the number of classes that need to be aggregated; cjRepresents the jth cluster; m isjRepresents a cluster CjThe cluster center of (a); dis denotes data point x and cluster center mjThe distance between them;
performing correlation analysis on the 24 acquisition items and the attention degree of the weapon equipment by means of a linear regression algorithm to obtain index weight of the acquisition items in the attention degree of the weapon equipment; the linear regression equation is:
wherein a and b are coefficients of the linear equation, and a and b are values solved by the following equations:
preferably, the six-dimensional weaponry attention model is embodied as follows:
(one), in the aspect of attack scope: the score of the index is derived from the resultant input of the weaponry during use; classifying various conventional weaponry according to the attack distance and by combining the median of the weaponry in the attack coverage range; classifying and counting the intermediate values of the weapon attack distance and the weapon attack coverage range, and dividing the criteria into:
MaxValue represents the maximum value of all weapon attack distances; MinValue represents the minimum of all weapons equipment attack distances; xi represents the intermediate value of the attack distance of certain weapon equipment; level represents the rounding of the calculation result, the value range is 1-n, and n represents the number of the divided levels;
(II) in the aspect of battle scenes: the score of the index is derived from actual combat use of the weaponry in the field outside the army, the score is 1-3 and belongs to low-level attention, the score is 4-6 and belongs to medium-level attention, and the score is 7-10 and belongs to high-level attention;
(III) in the aspect of maintenance difficulty: the score of the index is derived from the dynamic record of on-site maintenance and off-site entering overhaul of the weapon equipment, is used for tracking abnormal information describing the use of the weapon equipment during the service period, and belongs to low-level attention degree at 1-3, medium-level attention degree at 4-6 and high-level attention degree at 7-10;
(IV) in terms of equipment life: the score of the index is derived from statistics of the service life of weaponry in the field outside the military;
(V) in the aspect of operation difficulty: the score of the index is derived from the results of the soldier's experience with the weaponry equipment;
(VI) in the aspect of task difficulty: the score for the index is derived from the outcome of the performance of the weaponry in performing the task.
A system for task-based query and recommendation of weaponry knowledge-maps, the system comprising,
the weapon equipment entity label construction unit is used for extracting the basic information of the weapon equipment entity into a basic information label of the equipment entity through analyzing, filtering and cleaning the basic information of the weapon equipment entity;
the system comprises a weapon equipment capacity label construction unit, a weapon equipment capacity label analysis unit, a;
the weapon equipment knowledge map label construction unit is used for carrying out comprehensive analysis on a basic information label and an ability behavior label of a weapon equipment entity to carry out data modeling and then giving a cognitive label of the equipment ability;
and the entity tag construction unit based on the task recommendation equipment is used for constructing an equipment knowledge graph in the task process on the basis of the cognitive tag of the equipment capability, and further recommending the equipment with the tag to the completion target of the task so as to provide the development recommendation equipment of the task for use suggestion in the later period.
An electronic device, comprising: a memory and at least one processor;
wherein the memory stores computer-executable instructions;
the at least one processor executing the computer-executable instructions stored by the memory causes the at least one processor to perform the task-based weaponry knowledge-graph query and recommendation method described above.
A computer readable storage medium having stored therein computer executable instructions which, when executed by a processor, implement a method of task-based weaponry knowledge-maps querying and recommendation as described above.
The task-based weapon equipment knowledge graph query and recommendation method and system have the following advantages:
the method comprises the steps that (I) by means of advanced technologies of internet industries such as a knowledge graph and the like, and by combining warfare, war preparedness security, operational research and other computer science and theory concepts, a knowledge graph query and recommendation system based on task weapon equipment recommendation under the special condition of the military is constructed; according to the idea of selecting equipment to complete the task to recommending equipment to complete the task, through establishing a systematized equipment knowledge graph and a big data model and a layer-by-layer progressive equipment and task system, not only is the comprehensive cognition of dynamic insight on the weapon equipment capacity to the task demand capacity realized, but also the comprehensive equipment capacity association graph analysis and recommendation of targeted improvement opinions are realized; compared with the traditional equipment archive information system, the method not only provides a concept of pertinently dealing with a certain type of tasks from a knowledge map of the equipment capacity, but also realizes the support of a big data system of the equipment capacity;
secondly, data are extracted, cleaned and put into a big data analysis platform and a knowledge graph from different weaponry information systems; the records of basic information of weaponry in different systems are inconsistent, and data need to be extracted and cleaned firstly, so that data acquisition items meet the unified standard, and thus, the unified analysis and processing can be performed on all the existing systems;
thirdly, the data is extracted and cleaned from different equipment information recording systems and put into a big data analysis platform and a knowledge graph; the records of the strength information of the weapon equipment in different systems are inconsistent, and data needs to be extracted and cleaned firstly, so that data acquisition items meet the unified standard, and thus, the unified analysis and processing can be performed on all the existing systems;
fourthly, after the basic information label and the capability behavior label of the weapon equipment entity are comprehensively analyzed for data modeling, a cognitive label of the equipment capability is given out, and the accuracy and the integrity of the data are ensured;
(V) different influence factors are given to various abilities of equipment and tasks, the influence factors are variable and are mainly given according to historical experience values at present, data are more and more along with the construction of a big data platform and a knowledge graph in the later period, the experience values are corrected through a machine learning technology, a behavior model, a cognitive model and the like of the association relationship between the equipment and the tasks are continuously perfected, and a foundation is laid for realizing that the equipment is recommended to complete a certain type of tasks;
after the cognitive label of the weapon equipment capability is given, the information in the knowledge graph can be opened, so that different application systems use the information, further, a manager of the weapon equipment can comprehensively know the equipment, and the task and the use scene of the equipment are solidified, so that a continuous iterative closed loop is completed in a circulating mode, and the weapon equipment capability cognitive label becomes more and more intelligent and more accurate in the later use process.
Drawings
The invention is further described below with reference to the accompanying drawings.
FIG. 1 is a schematic diagram of a task-based weaponry knowledge-graph query and recommendation method.
Detailed Description
The task-based weaponry knowledge-map query and recommendation method and system of the present invention are described in detail below with reference to the figures and the detailed description of the preferred embodiments.
Example 1:
according to the task-based method for inquiring and recommending the knowledge graph of the weapon equipment, the three-element entity of the weapon equipment is used for integrally recognizing the weapon, and on the basis of gathering basic information and dynamic capacity information of the weapon equipment, the weapon equipment is subjected to omnibearing information display, analysis and mining through data standardization and visualization to construct the knowledge graph of the weapon entity; the invention provides the following in the field of military weapon equipment portrait: the method comprises the following steps of (1) a two-dimensional four-level equipment entity system, wherein the two-dimensional data is divided into two data processing dimensions of a data display dimension and a data analysis dimension, the four-level data is a four-level label system which is established as a ' weapon equipment entity label ', a weapon equipment capability label ', a weapon equipment knowledge map label ' and a task recommendation equipment entity label ', and a full-chain knowledge map system from equipment cognition to task support is realized, so that in a comprehensive way, the task-equipment knowledge map constructed by the method realizes three changes of equipment capacity and task guarantee:
(1) from the presentation of weaponry basic information to the recording of dynamic capability information. The traditional weapon equipment information is only static display of factory parameter data, dynamic data display and analysis of weapon equipment based on time series are realized by adding dynamic data such as weapon fire value, scoring evaluation, weapon maintenance and weapon use scene, and the triple entity of the weapon equipment is described more accurately and more timely.
(2) From "phenomenon" data of weaponry to "essence" characterization of weaponry. The basic information label and the dynamic capability label of the weapon equipment reflect the whole capability expression of the weapon equipment, the knowledge map entity representation of the weapon equipment reflects the deep description of the weapon equipment on the basis of big data analysis, and the key characteristics of the weapon equipment in the deep layer of the application process are reflected.
(3) From the presentation of independent information to the systematic modeling of four-level labels. The four-stage model of 'weapon equipment entity label-weapon equipment capability label-weapon equipment knowledge graph label-task recommendation equipment entity label-based task recommendation equipment entity label' is a thinking process from recognizing weapon equipment to recommending weapon equipment aiming at the task, and also embodies the cognitive transition of 'recommending equipment to complete certain type of task'.
The two-dimensional data processing method is different from the prior application scene that more internet customer images are applied to group screening, and the military uses the ternary entity group of the weapon equipment, which is focused on the screening of the whole variety and the query and analysis of all detailed information of the single weapon equipment. Therefore, the data presentation dimension focuses on presenting the weaponry information in a manner that is more adaptive to military managers, while the data analysis dimension focuses on implementing the background storage and knowledge graph building process by standardizing the data. Taking the "task capacity of air-sea integrated war" relationship as an example:
and performing visualization processing on the data of the display dimension according to item, and displaying each dimension which can be related to the equipment with sea-air capability. The analysis dimension is mainly processed from data standardization, and equipment capacity latitude which is possibly involved is completely standardized, so that although more storage space is occupied, the query and statistics speed of data is greatly improved. The display dimensions and analysis dimensions differ in the building thinking as shown in the following table:
the four levels are used for constructing a knowledge graph system of the task-based weapon equipment recommendation according to four levels of 'weapon equipment entity labels-weapon equipment capability labels-weapon equipment knowledge graph labels-task recommendation equipment entity labels'.
(1) Weapon equipment physical tag: the label category aims to ensure that equipment users can inquire various static data of equipment in a one-stop mode, and labels are inquired repeatedly.
(2) Weapon equipment capability tag: the dynamic entity capability is introduced, and the patent particularly introduces time series and abnormal behavior indexes aiming at the equipment triple entity portrait.
(3) Weapon equipment knowledge map label: through the organic combination of the basic information label and the equipment capability behavior label, cognitive labels such as capability scene images, equipment capability abnormal behavior scene images and the like are attached to the equipment capability.
(4) Recommending equipment entity labels based on the tasks: on the basis of the weapon equipment knowledge graph label, a task-equipment knowledge graph system with a task multi-equipment combination capability is provided in a targeted manner.
As shown in fig. 1, the method is as follows:
s1, constructing a weapon equipment entity label: extracting basic information of the weaponry entities into basic information tags of the weaponry entities through analysis, filtering and cleaning of the basic information of the weaponry entities;
s2, constructing a weaponry capability label: extracting the recent dynamic behaviors of the capabilities of the weaponry into capability behavior labels of the weaponry through analysis, filtering and cleaning of the recent dynamic behaviors of the capabilities of the weaponry;
s3, constructing a weapon equipment knowledge graph label: after carrying out comprehensive analysis on a basic information label and an ability behavior label of a weapon equipment entity for data modeling, giving a cognitive label of the equipment ability;
s4, constructing a task recommendation equipment-based entity label: on the basis of the cognitive label of the equipment capability, the equipment knowledge graph construction of the task process is added, and then the equipment with the label is recommended to the completion target of the task, so that the equipment recommendation suggestion is provided for the expansion of the task in the later period.
The basic information label of the first layer is based on the basic information of the existing weaponry. Because the informatization levels of different equipment management are different, and corresponding equipment basic information systems are different, according to the acquisition item standard specification of the national military standard, the basic information label is to clean and process heterogeneous data through an ETL tool, so that all data are subjected to two-stage specification according to display dimensions and analysis dimensions, and a standard and uniform data warehouse and equipment knowledge graph are created, thereby greatly improving the quality of the data, preparing the data for creating a model, avoiding the colloquial description, describing in an accurate language as much as possible, and if a large amount of manpower and material resources are input for filtering and cleaning, describing as follows: according to the requirements of the acquisition project, heterogeneous data is cleaned and loaded through an ETL tool, so that all data are subjected to two-stage specification according to display dimensions and analysis dimensions, existing equipment basic information systems are utilized, heterogeneous structures of equipment basic information systems in different scientific research institutions and armies are shielded, time and cost are greatly saved, and continuous operation of the existing systems is not influenced. The information development of military equipment is in a rapid development stage, so that the following problems exist in the existing equipment information: the heterogeneous problems of the system are more because the heterogeneous systems are stored in different systems; because of the development stage of digitalization and informatization, much information is incomplete. Therefore, a plurality of equipment information acquisition items need to be classified, each acquisition item is endowed with an influence factor, then a basic information label model of the equipment is trained and learned through a large amount of data, and finally a plurality of important acquisition items such as equipment attack, equipment use scenes and the like are selected from the plurality of basic information items for focusing attention. For example, regarding the acquisition item of the sea attack capability of weaponry, if an equipment type participates in the training task of sea attack many times, the equipment is given a basic information tag of sea attack capability. The equipment user can see the label, and does not need to go to each equipment information system to search the information of the equipment in the dimension.
The first layer of basic information labels is based on static information, while the use of the weaponry is a dynamically changing process, so that the second layer of weaponry capability labels is further proposed to perform big data analysis and deep description on the dynamic records of the weaponry at different periods; after the weaponry is arranged in the army, receiving a series of activities of using, transforming, maintaining and returning to a factory, generating new information every day, and updating the new information so as to better follow the capacity state change process of the weaponry; for example, the airplane equipment can find new changes through daily flight training, combat readiness duty, counterexercise and the like, different using modes and application scenes can timely reflect the capability condition of the equipment, and a track is formed in the time dimension. According to the change of the track, particularly the mutation information, the change of the current fighting capacity of the weapon equipment can be timely grasped. When one weapon equipment is not used for a long time or is frequently used in recent actions, setting thresholds for information of different equipment stages; when the threshold value is triggered, warning is given to the change of the fighting value of the equipment, and a weapon user or a maintainer is reminded to pay attention to the weapon equipment; therefore, the dynamic behavior of the weapon equipment is not a disordered record any more, but a capability trend of the weapon equipment depicts a track, and the improvement is facilitated.
The first layer basic information label and the second layer weapon equipment capability label respectively describe the labels of the weapon equipment from the static and dynamic aspects of the weapon equipment, the static and dynamic aspects complement each other in opposition and unity, and then the third layer weapon equipment capability label is further designed. The weapon equipment knowledge graph label is characterized in that different influence factors are respectively set for the static state and the dynamic state of weapon equipment, then big data is used for repeated training and modification, six dimensional indexes including combat scenes, attack tediousness, service life, operation difficulty, task difficulty and maintenance difficulty are selected from a plurality of acquisition items, and the weapon equipment is specifically and measurably analyzed; and then establishing a six-dimensional attention model of the weaponry by using the six-dimensional indexes, dividing the six-dimensional indexes of the combat scene, the attack tediousness, the service life, the operation difficulty, the task difficulty and the maintenance difficulty into 10 grades, and quantizing the indexes from 1 grade to 10 grades, wherein the larger the index value is, the more important attention is represented by the behavior of the weaponry in the dimension.
The establishment of the weapon equipment six-dimensional attention model is as follows:
(1) reducing the n-dimensional model into a 24-dimensional attention model by adopting a principal component analysis method in cluster analysis; the method comprises the following specific steps:
firstly, a K mean algorithm is adopted, the convergence condition is that the error square sum SSE is minimum, and the generated clusters are compact and independent by utilizing the criterion; the equation for the sum of squared errors SEE is as follows:
wherein k represents the number of classes that need to be aggregated; cjRepresents the jth cluster; m isjRepresents a cluster CjThe cluster center of (a); dis denotes data point x and cluster center mjThe distance between them;
performing correlation analysis on the 24 acquisition items and the attention degree of the weapon equipment by means of a linear regression algorithm to obtain index weight of the acquisition items in the attention degree of the weapon equipment; the linear regression equation is:
wherein a and b are coefficients of the linear equation, and a and b are values solved by the following equations:
(2) and selecting 6 acquisition items with highest correlation from the acquisition items and the attention models of the weaponry to construct a six-dimensional attention model of the weaponry. The six-dimensional weapon equipment attention model is specifically as follows:
(one), in the aspect of attack scope: the score of the index is derived from the resultant input of the weaponry during use; classifying various conventional weaponry according to the attack distance and by combining the median of the weaponry in the attack coverage range; classifying and counting the intermediate values of the weapon attack distance and the weapon attack coverage range, and dividing the criteria into:
MaxValue represents the maximum value of all weapon attack distances; MinValue represents the minimum of all weapons equipment attack distances; xi represents the intermediate value of the attack distance of certain weapon equipment; level is the rounding of the calculation result, and the value range is 1-10. If a weapon is equipped with multiple use modes and scenes, the attack distance is superposed according to a single level, and if the attack distance exceeds 10, the maximum value is 10. The scores were low level of attention at 1-3, medium level of attention at 4-6, and high level of attention at 7-10. For the weapon equipment with excellent performance such as artillery, tank and the like, the index is high, and for the weapon equipment with high level attention, the weapon equipment is worth focusing on the field and maintenance factory of the army, and the external high attack force of the army is maintained.
(II) in the aspect of battle scenes: the score of the index is derived from actual combat use of the weaponry in the field outside the army, the score is 1-3 and belongs to low-level attention, the score is 4-6 and belongs to medium-level attention, and the score is 7-10 and belongs to high-level attention; whether the weaponry is used in the army outfield or not is recorded; whether a combat performance record exists or not; whether a maintenance record exists or not; whether a battle damage telephone exists or not is recorded into the system, and further, the battle scene indexes of weaponry outside the army are influenced. In addition, aiming at the abnormal behavior of the weaponry in the army outfield use scene, a corresponding threshold value is set, if the frequency exceeds a certain frequency or the behaviors are subjected to intensive mutation, the score in the battle scene is low, the weaponry needs to be paid attention to in the aspect of army, manufacturers, maintenance factories and the like, the defect fault dispersion of the weaponry is timely made, and accidents are prevented.
(III) in the aspect of maintenance difficulty: the score of the index is derived from the dynamic record of on-site maintenance and off-site entering overhaul of the weapon equipment, is used for tracking abnormal information describing the use of the weapon equipment during the service period, and belongs to low-level attention degree at 1-3, medium-level attention degree at 4-6 and high-level attention degree at 7-10; the equipment guarantee and maintenance is an important concern of weaponry, the stable guarantee and maintenance capability is favorable for the service of the weaponry, and the weaponry with continuous fault problems and difficult maintenance and maintenance is not favorable for the use of army outfield. Any weapon equipment maintenance variation may cause equipment performance changes. In background data storage, all maintenance change information of weaponry is recorded into a big data analysis platform and a knowledge graph, and although a certain storage space is consumed, small changes of weaponry can be reflected. In the dimension of data analysis, a maintenance change model can be constructed for all maintenance changes of the weapon equipment, and the change model is stable or sudden; in the data display dimension, only the capability information of the changed weaponry is displayed, and redundant information is hidden, so that the analysis of a user is not influenced.
(IV) in terms of equipment life: the score of the index is derived from statistics of the service life of weaponry in the field outside the military; different weaponry have different lives, some are disposable consumables, and some can be used repeatedly. The service life of the equipment is an important factor for military use and modification of weaponry, and can also reflect the product characteristics of the weaponry and the habits of users, and even reflect the changes of equipment strength in recent time. The service life of the equipment is influenced by weather, task execution scenes, task execution grades, user capacity and the like, and after the weapons are used in the army outfield, the resource information, the use degree, the use time and other information of the corresponding weapons are recorded. Because the skills and hobbies of the weapon equipment users are stable in a period of time, the life duration information of the weapon equipment is normally distributed, when the using duration of the weapon equipment, the time period for executing tasks and the category for executing the tasks are suddenly changed, the score of the life index of the weapon equipment is lowered, and troops and maintainers need to pay more attention to the weapon equipment and guide the weapon equipment in time.
(V) in the aspect of operation difficulty: the score of the index is derived from the results of the soldier's experience with the weaponry equipment; in order to pay attention to the use dynamics of the weaponry in time, visualize the operation examination records of the equipment and be capable of knowing the fatality of the weaponry in the use process and implementing repair intervention in time, the system arranges the operation examination results and the maintenance implementation records of the weaponry, and forms a model of the operation difficulty index of the weaponry by means of the multi-year experience of equipment users such as military experts, and the like, classifies the operation difficulty of the weaponry, scores are 1-10, and people with higher scores indicate that the equipment is more difficult to operate and needs more important attention; then aiming at the operation difficulty of different grades of weaponry, the research and development force is reasonably distributed, and different operation transformation and production intervention schemes are implemented, so that the force of the weaponry can play a greater effect.
(VI) in the aspect of task difficulty: the score for the index is derived from the outcome of the performance of the weaponry in performing the task. The performance of the weapon equipment can be actually tested in the task process, the skill of the weapon equipment can be excavated, the function of the weapon equipment can be favorably exerted, the weapon user of an army can be favorably familiar with the equipment, and the method is an effective familiar way for the weapon equipment, so that the task process plays an important role in the use and transformation of the weapon equipment and also plays a greater role in a six-dimensional attention model. The assessment of the task difficulty index is also 1-10 points, and the higher the score is, the more important attention is needed. In the task execution process, the system can record the difficulty and completion condition of each task execution, the quality of equipment and other information, and through big data analysis of the dynamic information, the changes of weapons and equipment in different task processes can be reflected on a chart, so that a weapon manager can clearly see the condition of the weapons. For example, by comparing the duration of the mission process of the weapon in a certain day with the average duration of the previous days, if the decrease is much lower, the quality of the weapon decreases, the failure rate increases, and the threshold set by the system is exceeded, it can be considered that the mission capacity of the weapon has large fluctuation, the weapon needs to be focused, the weapon needs to be overhauled, the situation can be known in time, the cause of the abnormal fluctuation can be found, and the weapon can be guided to be transformed in time and effectively.
On the basis of a six-dimensional attention model, namely a battle scene, an attack range, a service life, operation difficulty, task difficulty and maintenance difficulty, a comprehensive attention index is finally established for different influence factors with different dimensions, the score of the index is 0-100, and the higher the score, the more important attention is needed. The score is more than 70 points, which indicates that the performance of the weapon equipment fluctuates greatly, the maintenance difficulty is high, all aspects of the equipment need to pay attention, the score is 40-70 points, which indicates that the performance of the weapon equipment may fluctuate greatly in some aspects, and several aspects need to be paid attention, and the score is less than 40 points, which indicates that the performance of the weapon is stable, the task ability is good, and light attention is needed.
In a six-dimensional attention model, the values of the influence factors occupy an important position. The default initial value of each influence factor set by the method is 0.01, and the value of each influence factor is optimized by means of machine learning knowledge and algorithm optimization through a plurality of rounds of training of a large amount of data. Taking a common weapon shooting distance as an example, when influence factors are initialized, the influence factors of each design influence factor are the same and are all 0.01, and through the technology of machine learning and the factor model transformation of the shooting distance of the weapon through multiple rounds, a new influence factor table and a rifle shooting distance influence factor comparison table are finally obtained, wherein the table comprises the following steps:
| classification of shots | Initial impact factor value | Final impact factor value |
| <＝10m | 0.01 | 0.64 |
| <＝50m | 0.01 | 0.64 |
| <＝100m | 0.01 | 0.63 |
| <＝150m | 0.01 | 0.59 |
| <＝200m | 0.01 | 0.58 |
| <＝250m | 0.01 | 0.55 |
| <＝300m | 0.01 | 0.55 |
| <＝350m | 0.01 | 0.55 |
| <＝400m | 0.01 | 0.55 |
| <＝500m | 0.01 | 0.54 |
| <＝800m | 0.01 | 0.54 |
| <＝1000m | 0.01 | 0.53 |
| <＝1500m | 0.01 | 0.52 |
| <＝2000m | 0.01 | 0.39 |
| <＝2500m | 0.01 | 0.31 |
| <＝3000m | 0.01 | 0.01 |
| <＝3500m | 0.01 | 0.01 |
| … | 0.01 | 0.01 |
From the above table, it can be seen that most rifles have a range of about 100m, which corresponds to a high influence factor, and the result is consistent with the research and development design.
The four-level method system of 'weapon equipment entity label-weapon equipment capability label-weapon equipment knowledge map label-task recommendation equipment entity label' in the invention is a closed loop with progressive and repeated iteration, and the weapon equipment label given by the model is more and more accurate with the help of big data analysis and machine learning training, and the provided equipment improvement suggestion is more and more reasonable.
Example 2:
the invention relates to a system for querying and recommending a knowledge graph of weapon equipment based on tasks, which comprises,
the weapon equipment entity label construction unit is used for extracting the basic information of the weapon equipment entity into a basic information label of the equipment entity through analyzing, filtering and cleaning the basic information of the weapon equipment entity;
the system comprises a weapon equipment capacity label construction unit, a weapon equipment capacity label analysis unit, a;
the weapon equipment knowledge map label construction unit is used for carrying out comprehensive analysis on a basic information label and an ability behavior label of a weapon equipment entity to carry out data modeling and then giving a cognitive label of the equipment ability;
and the entity tag construction unit based on the task recommendation equipment is used for constructing an equipment knowledge graph in the task process on the basis of the cognitive tag of the equipment capability, and further recommending the equipment with the tag to the completion target of the task so as to provide the development recommendation equipment of the task for use suggestion in the later period.
The task-based weapon equipment knowledge graph query and recommendation system has the following working processes:
(1) the weapon equipment physical label construction unit collects, cleans and filters the basic information of criminals from the existing file card information system or a special collected information management system, and stores the data into a weapon equipment portrait data warehouse.
Because the informatization levels of different troops are different, the gear card information systems of different troops are different, and in order to achieve the standardization and unification, all acquisition items are screened and cleaned according to the standard specification of the acquisition items of the military.
(2) The weapon equipment capability label construction unit cleans and standardizes the dynamic information of the weapon equipment collected from the task system and other systems, and labels the dynamic behavior of the weapon performance and function.
The weapon equipment capability label construction unit realizes weapon dynamic data display and analysis based on time series by adding dynamic data such as score assessment, maintenance assessment, function use and the like, and the weapon equipment is more accurately and timely described.
(3) After the basic information labels and the dynamic behavior labels of the weaponry are processed, the knowledge map label construction unit of the weaponry can form the cognitive labels of the weaponry.
The weapon equipment knowledge map label construction unit is based on a weapon equipment entity label construction unit and a weapon equipment capability label construction unit. A comprehensive evaluation was performed on the results of the first two steps. Different indexes are endowed with different weights, a series of influence factors are finally determined through machine learning and training of big data, and processing of cognitive tags of weaponry is completed. The main algorithm used is as follows,
wherein n isiRefers to the value of the ith basic information tag or dynamic behavior tag; x is the number ofiThe specific gravity of the corresponding label in the whole model is shown, and I is the sum calculated by the index model; different influence factors are assigned to different indexes. These impact factors need to be adjusted according to different application scenarios. In the present embodiment of the present invention,a total of 24 major classes are established, for a total of 180 small terms, with a default value of 0.01 for each impact factor. In order to enable the model to be more accurate, the value range of 0.01-1 is given to all the influence factors in the model, then the model result corresponding to each value is calculated through programming, and after numerous model results are screened, an influence factor vector which is well matched in all aspects is selected.
(4) After the cognitive label of the weapon is processed, the use and modification strategy label of the weapon is given by the task recommendation equipment entity label building unit according to the cognitive label of the weapon equipment.
The four-stage method system of 'weapon equipment entity label-weapon equipment capability label-weapon equipment knowledge map label-based task recommendation equipment entity label' in the embodiment of the invention is dynamic closed-loop iteration, and through repeated big data analysis and machine learning training, the weapon cognitive label that can be given by the model is more and more accurate, the provided assistant and education suggestion is more and more accurate, the comprehensive understanding and accurate transformation of troops on weapon equipment are promoted, and the improvement of 'permanent solution safety view' to 'permanent solution safety view' is promoted.
Example 3:
an embodiment of the present invention further provides an electronic device, including: a memory and at least one processor;
wherein the memory stores computer-executable instructions;
the at least one processor executing the memory-stored computer-executable instructions cause the at least one processor to perform a method of task-based querying and recommending weapons gear knowledge-maps as in any of the embodiments.
Example 4:
embodiments of the present invention also provide a computer-readable storage medium having stored thereon a plurality of instructions, which are loaded by a processor, to cause the processor to perform a method for task-based querying and recommending a weaponry knowledge-graph according to any of the embodiments of the present invention. Specifically, a system or an apparatus equipped with a storage medium on which software program codes that realize the functions of any of the above-described embodiments are stored may be provided, and a computer (or a CPU or MPU) of the system or the apparatus is caused to read out and execute the program codes stored in the storage medium.
In this case, the program code itself read from the storage medium can realize the functions of any of the above-described embodiments, and thus the program code and the storage medium storing the program code constitute a part of the present invention.
Examples of storage media that can be used to provide program code include floppy disks, hard disks, magneto-optical disks, optical disks (e.g., CD-ROMs, CD-R, CD-RWs, DVD-ROMs, DVD-R task-based methods and systems M, DVD-RWs, DVD + RWs), magnetic tape, non-volatile memory cards, and ROMs. Alternatively, the program code may be downloaded from a server computer via a communications network.
Further, it should be clear that the functions of any one of the above-described embodiments may be implemented not only by executing the program code read out by the computer, but also by causing an operating system or the like operating on the computer to perform a part or all of the actual operations based on instructions of the program code.
Further, it is to be understood that the program code read out from the storage medium is written to a memory provided in an expansion board inserted into the computer or to a memory provided in an expansion unit connected to the computer, and then causes a CPU or the like mounted on the expansion board or the expansion unit to perform part or all of the actual operations based on instructions of the program code, thereby realizing the functions of any of the above-described embodiments.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. A task-based method for inquiring and recommending a knowledge graph of weapon equipment is characterized in that a weapon equipment ternary entity is used for integrally recognizing weapons, and omnibearing information display, analysis and mining are carried out on the weapon equipment through data standardization and visualization on the basis of gathering basic information and dynamic capacity information of the weapon equipment to construct the knowledge graph of the weapon entity; the method comprises the following specific steps:
s1, constructing a weapon equipment entity label: extracting basic information of the weaponry entities into basic information tags of the weaponry entities through analysis, filtering and cleaning of the basic information of the weaponry entities;
s2, constructing a weaponry capability label: extracting the recent dynamic behaviors of the capabilities of the weaponry into capability behavior labels of the weaponry through analysis, filtering and cleaning of the recent dynamic behaviors of the capabilities of the weaponry;
s3, constructing a weapon equipment knowledge graph label: after carrying out comprehensive analysis on a basic information label and an ability behavior label of a weapon equipment entity for data modeling, giving a cognitive label of the equipment ability;
s4, constructing a task recommendation equipment-based entity label: on the basis of the cognitive label of the equipment capability, the equipment knowledge graph construction of the task process is added, and then the equipment with the label is recommended to the completion target of the task, so that the equipment recommendation suggestion is provided for the expansion of the task in the later period.
2. The method of claim 1, wherein the basic information labels are obtained by washing heterogeneous data with an ETL tool, so that all data are subjected to two-level specification according to display dimension and analysis dimension, and a unified data warehouse and equipment knowledge graph is created.
3. The method for task-based weaponry knowledge-graph query and recommendation according to claim 1, wherein weaponry capability labels are big data analysis and deep delineation of weaponry dynamic records at different times;
after the weaponry is arranged in the army, receiving a series of activities of using, transforming, maintaining and returning to a factory, generating new information every day, and updating the new information so as to better follow the capacity state change process of the weaponry;
when one weapon equipment is not used for a long time or is frequently used in recent actions, setting thresholds for information of different equipment stages; when the threshold value is triggered, a warning is sent to the change of the fighting value of the equipment, and a weapon user or a maintainer is reminded to pay attention to the weapon equipment.
4. The method for querying and recommending a mission-based weaponry knowledge-graph according to any of claims 1-3, wherein the weaponry knowledge-graph labels are obtained by setting different influence factors for the static state and the dynamic state of weaponry, repeatedly training and modifying the weaponry knowledge-graph labels by using big data, selecting six dimensional indexes of combat scene, attack tediousness, service life, operation difficulty, mission difficulty and maintenance difficulty from a plurality of collected items, and specifically and measurably analyzing the weaponry; and then establishing a six-dimensional attention model of the weaponry by using the six-dimensional indexes, dividing the six-dimensional indexes of the combat scene, the attack tediousness, the service life, the operation difficulty, the task difficulty and the maintenance difficulty into 10 grades, and quantizing the indexes from 1 grade to 10 grades, wherein the larger the index value is, the more important attention is represented by the behavior of the weaponry in the dimension.
5. The method for querying and recommending a mission-based weaponry knowledge-graph as claimed in claim 4, wherein the establishment of a weaponry six-dimensional attention model is as follows:
(1) reducing the n-dimensional model into a 24-dimensional attention model by adopting a principal component analysis method in cluster analysis;
(2) and selecting 6 acquisition items with highest correlation from the acquisition items and the attention models of the weaponry to construct a six-dimensional attention model of the weaponry.
6. The method for querying and recommending a mission-based weaponry knowledge-graph according to claim 5, wherein the n-dimensional model is reduced to a 24-dimensional attention model by using a principal component analysis method in cluster analysis as follows:
firstly, a K mean algorithm is adopted, the convergence condition is that the error square sum SSE is minimum, and the generated clusters are compact and independent by utilizing the criterion; the equation for the sum of squared errors SEE is as follows:
wherein k represents the number of classes that need to be aggregated; cjRepresents the jth cluster; m isjRepresents a cluster CjThe cluster center of (a); dis denotes data point x and cluster center mjThe distance between them;
performing correlation analysis on the 24 acquisition items and the attention degree of the weapon equipment by means of a linear regression algorithm to obtain index weight of the acquisition items in the attention degree of the weapon equipment; the linear regression equation is:
wherein a and b are coefficients of the linear equation, and a and b are values solved by the following equations:
7. the method for task-based weaponry knowledge-graph query and recommendation according to claim 5, wherein the six-dimensional weaponry attentiveness model is specified as follows:
(one), in the aspect of attack scope: the score of the index is derived from the resultant input of the weaponry during use; classifying various conventional weaponry according to the attack distance and by combining the median of the weaponry in the attack coverage range; classifying and counting the intermediate values of the weapon attack distance and the weapon attack coverage range, and dividing the criteria into:
MaxValue represents the maximum value of all weapon attack distances; MinValue represents the minimum of all weapons equipment attack distances; xi represents the intermediate value of the attack distance of certain weapon equipment; level represents the rounding of the calculation result, the value range is 1-n, and n represents the number of the divided levels;
(II) in the aspect of battle scenes: the score of the index is derived from actual combat use of the weaponry in the field outside the army, the score is 1-3 and belongs to low-level attention, the score is 4-6 and belongs to medium-level attention, and the score is 7-10 and belongs to high-level attention;
(III) in the aspect of maintenance difficulty: the score of the index is derived from the dynamic record of on-site maintenance and off-site entering overhaul of the weapon equipment, is used for tracking abnormal information describing the use of the weapon equipment during the service period, and belongs to low-level attention degree at 1-3, medium-level attention degree at 4-6 and high-level attention degree at 7-10;
(IV) in terms of equipment life: the score of the index is derived from statistics of the service life of weaponry in the field outside the military;
(V) in the aspect of operation difficulty: the score of the index is derived from the results of the soldier's experience with the weaponry equipment;
(VI) in the aspect of task difficulty: the score for the index is derived from the outcome of the performance of the weaponry in performing the task.
8. A system for task-based query and recommendation of weaponry knowledge-maps, the system comprising,
the weapon equipment entity label construction unit is used for extracting the basic information of the weapon equipment entity into a basic information label of the equipment entity through analyzing, filtering and cleaning the basic information of the weapon equipment entity;
the system comprises a weapon equipment capacity label construction unit, a weapon equipment capacity label analysis unit, a;
the weapon equipment knowledge map label construction unit is used for carrying out comprehensive analysis on a basic information label and an ability behavior label of a weapon equipment entity to carry out data modeling and then giving a cognitive label of the equipment ability;
and the entity tag construction unit based on the task recommendation equipment is used for constructing an equipment knowledge graph in the task process on the basis of the cognitive tag of the equipment capability, and further recommending the equipment with the tag to the completion target of the task so as to provide the development recommendation equipment of the task for use suggestion in the later period.
9. An electronic device, comprising: a memory and at least one processor;
wherein the memory stores computer-executable instructions;
the at least one processor executing the memory-stored computer-executable instructions to cause the at least one processor to perform the method of task-based weaponry knowledge-maps querying and recommending of claim 1 through 7.
10. A computer-readable storage medium having computer-executable instructions stored thereon that, when executed by a processor, perform the method for task-based querying and recommending weaponry knowledge-maps of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011136229.XA CN112241459A (en) | 2020-10-22 | 2020-10-22 | Task-based weapon equipment knowledge graph query and recommendation method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011136229.XA CN112241459A (en) | 2020-10-22 | 2020-10-22 | Task-based weapon equipment knowledge graph query and recommendation method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112241459A true CN112241459A (en) | 2021-01-19 |
Family
ID=74168188
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011136229.XA Pending CN112241459A (en) | 2020-10-22 | 2020-10-22 | Task-based weapon equipment knowledge graph query and recommendation method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112241459A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108399190A (en) * | 2018-01-24 | 2018-08-14 | 山东中磁视讯股份有限公司 | A kind of panorama image space method applied to prison convict |
| CN110909153A (en) * | 2019-10-22 | 2020-03-24 | 中国船舶重工集团公司第七0九研究所 | Knowledge graph visualization method based on semantic attention model |
| KR102079289B1 (en) * | 2019-04-23 | 2020-04-07 | 주식회사 비닛 | Wine recommendation system and method |
| CN111078868A (en) * | 2019-06-04 | 2020-04-28 | 中国人民解放军92493部队参谋部 | Knowledge graph analysis-based equipment test system planning decision method and system |
| CN111309925A (en) * | 2020-02-10 | 2020-06-19 | 同方知网(北京)技术有限公司 | Knowledge graph construction method of military equipment |
-
2020
- 2020-10-22 CN CN202011136229.XA patent/CN112241459A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108399190A (en) * | 2018-01-24 | 2018-08-14 | 山东中磁视讯股份有限公司 | A kind of panorama image space method applied to prison convict |
| KR102079289B1 (en) * | 2019-04-23 | 2020-04-07 | 주식회사 비닛 | Wine recommendation system and method |
| CN111078868A (en) * | 2019-06-04 | 2020-04-28 | 中国人民解放军92493部队参谋部 | Knowledge graph analysis-based equipment test system planning decision method and system |
| CN110909153A (en) * | 2019-10-22 | 2020-03-24 | 中国船舶重工集团公司第七0九研究所 | Knowledge graph visualization method based on semantic attention model |
| CN111309925A (en) * | 2020-02-10 | 2020-06-19 | 同方知网(北京)技术有限公司 | Knowledge graph construction method of military equipment |
Non-Patent Citations (1)
| Title |
|---|
| 丁君怡等: ""基于开源数据的武器装备知识图谱构建方法研究"", 《指挥控制与仿真》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Akgün et al. | Solving an ammunition distribution network design problem using multi-objective mathematical modeling, combined AHP-TOPSIS, and GIS | |
| McClendon et al. | Using machine learning algorithms to analyze crime data | |
| US11093568B2 (en) | Systems and methods for content management | |
| Lemaire et al. | A survey on supervised classification on data streams | |
| Bentley et al. | Fuzzy darwinian detection of credit card fraud | |
| De Sá et al. | Mining association rules for label ranking | |
| CN114638234B (en) | Big data mining method and system applied to online business handling | |
| CN111557011A (en) | Enterprise yield-breaking prediction system and operation method thereof | |
| Li et al. | Battle damage-oriented spare parts forecasting method based on wartime influencing factors analysis and ε-support vector regression | |
| CN112241459A (en) | Task-based weapon equipment knowledge graph query and recommendation method and system | |
| CN108510393A (en) | Intelligence based on big data, which is thrown, cares for system and method | |
| Kishore et al. | Prediction of student academic progression: A case study on Vignan University | |
| Smith | Mission dependency index of air force built infrastructure: Knowledge discovery with machine learning | |
| Killicoat | Weaponomics: The Economics of Small Arms. | |
| CN115204457A (en) | Loan default risk prediction method based on graph attention network | |
| Legoui et al. | Equilibrium optimizer and henry gas solubility optimization algorithms for feature selection: comparison study | |
| Jokic et al. | Selection of Rifle Caliber in Rearming Process of the Serbian Army | |
| Nieves et al. | Finding patterns of terrorist groups in Iraq: A knowledge discovery analysis | |
| Haley et al. | Cluster analysis of deep embeddings in real-time strategy games | |
| Belue | Multilayer perceptrons for classification | |
| Li et al. | Forecasting firm risk in the emerging market of China with sequential optimization of influence factors on performance of case‐based reasoning: an empirical study with imbalanced samples | |
| Chaudhary | Data mining system, functionalities and applications: a radical review | |
| Iorio et al. | A combined clustering and multi-criteria approach for portfolio selection | |
| Muniandy et al. | Deep learning in detecting financial statement fraud: an application of deep neural network (DNN) | |
| Gerdelius et al. | Detecting Fraudulent User Behaviour: A Study of User Behaviour and Machine Learning in Fraud Detection |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210119 |