A method, device, equipment and storage medium for aircraft intention recognition based on knowledge graph

Abstract

The present application discloses a method for aircraft intention recognition based on a knowledge graph, the method comprising: obtaining knowledge data of aircraft intention, the knowledge data comprising entities, relationships, attributes and / or events, processing the knowledge data, eliminating contradictions and ambiguities therein, and merging duplicate knowledge data; generating a knowledge graph of target intention based on the processed knowledge data; obtaining the intention data of the target aircraft based on the knowledge graph of the target intention. The present application also discloses an aircraft intention recognition device based on a knowledge graph, corresponding equipment and storage medium. The present application can realize intelligent recognition of the intention of the target aircraft in a complex environment.

Description

Technical Field

[0001] The present application relates to the technical field of target intention recognition, and more specifically, to a method, device, equipment and storage medium for aircraft intention recognition based on knowledge graph. Background Art

[0002] Due to the increasing complexity and information asymmetry of the aircraft operating environment, it is difficult to judge and identify the other party's target intentions based on personal experience alone, so an intelligent recognition prediction method is needed.

[0003] Traditional methods usually rely on rule-based or classical machine learning techniques to achieve this goal. However, these methods usually face the following major problems:

[0004] Data silo problem: In a multi-source heterogeneous intelligence information environment, data is usually isolated and scattered, which makes it difficult to integrate and share information. Such "data silo" problems lead to inefficient information extraction and utilization, and also reduce the accuracy and efficiency of intent recognition.

[0005] Lack and insufficiency of semantic information: Traditional methods often ignore or fail to make adequate use of semantic information, making it difficult to understand and analyze information, reducing the accuracy and reliability of intent recognition, and limiting the reasoning and analysis capabilities of the model. Summary of the invention

[0006] In response to at least one defect or improvement need in the prior art, the present application provides an aircraft intention recognition method, apparatus, device and storage medium based on a knowledge graph, which can solve the problems existing in the above-mentioned prior art.

[0007] To achieve the above objectives, according to a first aspect of the present application, a method for aircraft intention recognition based on a knowledge graph is provided, the method comprising:

[0008] Acquire knowledge data of aircraft intent, the knowledge data including entities, relationships, attributes and / or events, process the knowledge data, eliminate contradictions and ambiguities therein, and merge duplicate knowledge data;

[0009] Generate a knowledge graph of target intent based on the processed knowledge data;

[0010] Based on the knowledge graph of target intent, the intent data of the target aircraft is obtained.

[0011] Furthermore, in the above-mentioned knowledge graph-based aircraft intention recognition method, the knowledge graph of the target intention includes a pattern layer and a data layer;

[0012] The model layer is built on the data layer and describes entities, relationships, attributes and / or events in a formal way;

[0013] The data layer contains the associations and relationships between instances and is the instance representation of the model layer.

[0014] Furthermore, in the above-mentioned aircraft intention recognition method based on knowledge graph, the pattern layer includes an attribute layer, a feature layer and an intention layer;

[0015] The attribute layer includes target-specific attribute information, which is a parameter used for target intent analysis;

[0016] The feature layer uses features as basic nodes, and the node information of the feature layer includes feature types and feature parameters;

[0017] The node information of the intent layer includes the inclusion relationship, relationship value domain and relationship constraint between target entity intents.

[0018] Furthermore, in the above-mentioned aircraft intention recognition method based on knowledge graph, the knowledge graph based on target intention obtains the intention data of the target aircraft, including:

[0019] In the knowledge graph of target intent, the target aircraft is connected to the entity through a direct relationship or an indirect relationship path;

[0020] The direct relationship or indirect relationship path is used as the attribute of the target aircraft, and the intention data of the target aircraft is obtained through the embedded propagation method.

[0021] Furthermore, in the above-mentioned knowledge graph-based aircraft intention recognition method, the direct relationship or indirect relationship path is used as the attribute of the target aircraft, specifically including:

[0022] The propagation information of the target entity is represented by defining a set of entity and relationship paths

[0023]

[0024] in For the initial set ;|| means putting relation r into the relation path A new path is formed behind represents the relationship path propagating from the target entity i to the jth hop;

[0025] The attention weight of the target-relation path is

[0026]

[0027] The attention weight of the target-entity path is

[0028]

[0029] The final weight obtained by normalization is

[0030]

[0031] Perform a linear combination of the weight information to obtain

[0032]

[0033] Get the set of all L-hop entity embedding information

[0034] .

[0035] Furthermore, in the above-mentioned aircraft intention recognition method based on knowledge graph, the intention data of the target aircraft is obtained by the embedding propagation method, and the intention data includes the probability score of the target intention, specifically including:

[0036] Will gather Inside Perform a sum operation to obtain a new embedding :

[0037]

[0038] Combine the collaborative information between targets into the target embedding to obtain the embedding of target u

[0039]

[0040] Where T(u) represents the set of intentions that have interacted with target u;

[0041] right and Perform inner product operation to obtain the probability score of predicting target intention

[0042] .

[0043] According to the second aspect of the present application, there is also provided an aircraft intention recognition device based on a knowledge graph, which includes:

[0044] A processing module, used to obtain knowledge data of aircraft intent, the knowledge data including entities, relationships, attributes and / or events, process the knowledge data, eliminate contradictions and ambiguities therein, and merge duplicate knowledge data;

[0045] A generation module is used to generate a knowledge graph of target intent based on the processed knowledge data;

[0046] The acquisition module is used to obtain the intention data of the target aircraft based on the knowledge graph of the target intention.

[0047] According to the third aspect of the present application, there is also provided an aircraft intention recognition device based on a knowledge graph, which includes at least one processing unit and at least one storage unit, wherein the storage unit stores a computer program, and when the computer program is executed by the processing unit, the processing unit performs the steps of the above method.

[0048] According to the fourth aspect of the present application, a storage medium is also provided, which stores a computer program that can be executed by an aircraft intent recognition device based on a knowledge graph. When the computer program runs on the aircraft intent recognition device based on the knowledge graph, the aircraft intent recognition device based on the knowledge graph executes the steps of the above method.

[0049] According to the fifth aspect of the present application, a computer program product is also provided, including a computer program, which, when executed by a processor, implements the steps of the above-mentioned knowledge graph-based aircraft intention recognition method.

[0050] In general, the above technical solutions conceived by the present application can achieve the following beneficial effects compared with the prior art:

[0051] (1) This application provides a method for aircraft intention recognition based on knowledge graph, which utilizes knowledge graph embedding learning technology to organize multi-source heterogeneous environmental intelligence data through target intention knowledge graph, integrate and extract information, improve the accuracy and efficiency of intention recognition, make full use of semantic analysis, improve the reasoning and analysis capabilities of the target intention recognition model, form multi-dimensional information understanding capabilities across time and space, across tasks, and across entities, and realize intelligent recognition of target aircraft intentions in complex environments.

[0052] (2) The present application provides a method for aircraft intention recognition based on knowledge graph, which adopts data fusion technology and can integrate data from multiple sources to form a rich and multi-dimensional data set, thereby providing a basis for deeper analysis. At the same time, through semantic technology, it can enhance the semantic level of data and achieve deeper data understanding and analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0054] Figure 1 A schematic diagram of a flow chart of a method for aircraft intention recognition based on a knowledge graph provided in an embodiment of the present application;

[0055] Figure 2 A schematic diagram of the knowledge graph construction process provided in the embodiment of the present application;

[0056] Figure 3 A schematic diagram of the data layer structure provided in the embodiment of the present application;

[0057] Figure 4 A schematic diagram of the mode layer structure provided in an embodiment of the present application;

[0058] Figure 5 A schematic diagram of the structure of an aircraft intention recognition device based on a knowledge graph provided in an embodiment of the present application. DETAILED DESCRIPTION

[0059] In order to make the purpose, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. In addition, the technical features involved in each embodiment of the present application described below can be combined with each other as long as they do not conflict with each other.

[0060] The terms "first", "second", "third", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0061] Figure 1 is a flow chart of a method for aircraft intention recognition based on a knowledge graph provided in an embodiment of the present application. Figure 1 As shown, the present application provides a method for aircraft intention recognition based on knowledge graph, including:

[0062] 101. Acquire knowledge data of aircraft intent, wherein the knowledge data includes entities, relationships, attributes and / or events, process the knowledge data, eliminate contradictions and ambiguities therein, and merge duplicate knowledge data.

[0063] Specifically, based on the visual modeling method, knowledge representation technology can be used to define concepts, attributes and / or rules related to aircraft, clarify the semantic associations between concepts, determine the semantic associations between concepts, attributes and / or rules, and realize the construction of a multi-level domain knowledge system.

[0064] The knowledge data of aircraft intent includes entities, relationships, attributes and / or events, where entities represent entities of target aircraft, relationships include inclusion relationships, relationship value ranges, relationship constraints, etc. between target entity intents, and attributes and / or events are attribute information inherent in the target itself. These attributes can be parameters used for target intent analysis, such as: the aircraft body has maximum speed, minimum speed, cruising speed and cruising altitude, maximum acceleration, maximum deceleration, maximum overload climb, maximum overload dive, etc.

[0065] Through entity recognition, relationship extraction and other knowledge extraction related technologies, knowledge elements such as entities, relationships, attributes, events, etc. are automatically extracted from structured, semi-structured and unstructured data. Entity disambiguation technology and entity linking technology are used to eliminate contradictions and ambiguities between knowledge objects such as entities, attributes, relationships, events, etc. Knowledge merging technology is used to merge duplicate knowledge. Knowledge reasoning, knowledge calculation, knowledge completion and other technologies are used to start from the existing entity relationship data in the knowledge base, and through reasoning and calculation, new association relationships between entities are established, thereby expanding and updating the knowledge network and obtaining processed knowledge data.

[0066] 102. Generate a knowledge graph of target intent based on the processed knowledge data.

[0067] Specifically, the process of constructing a knowledge graph is as follows: Figure 2 As shown, the processed knowledge data is obtained, including basic equipment information data, event text data, situation data and rule data, etc., and knowledge extraction is performed on the knowledge data, including entity extraction, relationship extraction, attribute extraction and time extraction, etc., and knowledge fusion is performed on the data after knowledge extraction, including entity disambiguation, entity linking and knowledge merging, etc. After merging duplicate data, knowledge update is performed, including knowledge calculation, knowledge reasoning and knowledge completion, to obtain the knowledge graph of the target intention.

[0068] 103. Based on the knowledge graph of target intention, obtain the intention data of the target aircraft.

[0069] Specifically, environmental intelligence data is obtained, and a target activity factor association network is constructed based on the target intention knowledge graph and environmental intelligence data, and the target intention knowledge graph is applied, such as Figure 2 As shown, intention recognition and result feedback are performed to obtain the intention recognition result of the target aircraft.

[0070] In the process of applying the target intention knowledge graph, new requirements and new concepts will be fed back to guide the supplement and improvement of the cognitive framework in the target intention field. Based on the intention recognition and result feedback of the target aircraft, combined with the acquisition of new requirements and new concepts, the processed data can be represented by knowledge, and its attributes, concept layers and rules can be defined. Knowledge extraction can be performed again, and the knowledge graph of target intention can be reconstructed based on historical data to achieve continuous recognition of target intention in complex environments.

[0071] In order to facilitate the use of real-time target intent recognition tasks, knowledge graph embedding learning technology can be used to obtain the implicit semantic information of entities and relationships. Each target in the target intent knowledge graph will have other entities connected to it through direct relationships or indirect relationship paths. These relationships mainly include affiliation, inclusion, and collaboration. These entities can be regarded as attributes of the target. Combined with the inherent attribute information and target state information of the target, it can reflect the target's intention tendency to a certain extent. This knowledge information can be utilized through the embedding propagation method.

[0072] The knowledge graph-based aircraft intention recognition method provided in the embodiment of the present application utilizes knowledge graph embedding learning technology to organize multi-source heterogeneous environmental intelligence data through the target intention knowledge graph, integrate and extract information, thereby improving the accuracy and efficiency of intention recognition, making full use of semantic analysis, and improving the reasoning and analysis capabilities of the target intention recognition model, forming a multi-dimensional information understanding capability across time and space, across tasks, and across entities, thereby realizing intelligent recognition of target aircraft intentions in complex environments.

[0073] Optionally, in the aircraft intention recognition method based on knowledge graph provided in the embodiment of the present application, the knowledge graph of target intention includes a pattern layer and a data layer;

[0074] The model layer is built on top of the data layer and describes entities, relationships, attributes and / or events in a formal way;

[0075] The data layer contains the associations and relationships between instances and is the instance representation of the model layer.

[0076] Specifically, in terms of logical architecture, the target intent knowledge graph can be divided into two aspects: the pattern layer and the data layer. The pattern layer is built on the data layer and describes pattern information such as related entities, relationships, attributes and / or events in a formal way. The pattern layer includes the attribute layer, feature layer and intent layer.

[0077] Among them, the attribute layer includes entity attributes, which may include entity-inherent attributes such as type, maximum speed, maximum acceleration, etc.; the feature layer includes motion state, entity attributes and sensor state. The motion state includes speed, steering angle, distance and other states. The entity attributes include type, maximum speed, maximum acceleration and other attributes. The sensor state includes uncontrolled radar state, controlled radar state, anti-electric equipment state and other states. The intention layer includes the intention type of the aircraft, which may include reconnaissance intention, attack intention, support intention and other intentions.

[0078] The data layer contains many associations between instances and is an instance representation of the pattern layer. For example, the data layer can include various data of the target aircraft, such as Figure 3 As shown, it may include standard data such as the aircraft's behavioral array radar data, speed data, acceleration data, etc., and it may also include dynamic data such as the aircraft's altitude data, attitude data, radio data, and speed data when penetrating.

[0079] Optionally, in the aircraft intention recognition method based on knowledge graph provided in the embodiment of the present application, the pattern layer includes an attribute layer, a feature layer and an intention layer;

[0080] The attribute layer includes the target's inherent attribute information, which is the parameter used for target intention analysis;

[0081] The feature layer uses features as basic nodes, and the node information of the feature layer includes feature types and feature parameters;

[0082] The node information of the intent layer includes the inclusion relationship, relationship value domain and relationship constraints between the target entity intents.

[0083] Specifically, the attribute layer includes the target's inherent attribute information, which is the parameters used for target intent analysis, including the aircraft's maximum speed, minimum speed, cruising speed and cruising altitude, maximum acceleration, maximum deceleration, maximum overload climb, maximum overload dive, etc.

[0084] The feature layer uses various features as basic nodes. The node information mainly includes feature type and feature parameters, such as sensor status, maneuver type, environmental status, our target status, etc.

[0085] The node information of the intent layer mainly includes the inclusion relationship, relationship value range, relationship constraints, etc. between the target entity intentions.

[0086] like Figure 4As shown, it is an example of the construction of the attribute layer, feature layer, and intention layer given in the embodiment of the present application, wherein the attribute layer includes entity attributes, which may include entity-specific attributes such as type, maximum speed, and maximum acceleration; the feature layer includes motion state, entity attributes, and sensor state, the motion state includes states such as speed, steering angle, and distance, the entity attributes include attributes such as type, maximum speed, and maximum acceleration, and the sensor state includes states such as uncontrolled radar state, controlled radar state, and anti-electric equipment state; the intention layer includes the intention type of the aircraft, which may include intentions such as reconnaissance intention, attack intention, and support intention. Among them, the relationship constraints may include constraints such as intention type constraints, attack distance constraints, and speed constraints; the intention analysis may include intention analysis such as speed analysis, range distance analysis, and maneuver type analysis.

[0087] Optionally, the aircraft intention recognition method based on the knowledge graph provided in the embodiment of the present application obtains the intention data of the target aircraft based on the knowledge graph of the target intention, including:

[0088] In the knowledge graph of target intent, the target aircraft is connected to the entity through a direct relationship or an indirect relationship path;

[0089] The direct or indirect relationship paths are taken as the attributes of the target aircraft, and the intention data of the target aircraft is obtained through the embedding propagation method.

[0090] Optionally, the aircraft intention recognition method based on the knowledge graph provided in the embodiment of the present application uses the direct relationship or indirect relationship path as the attribute of the target aircraft, specifically including:

[0091] The propagation information of the target entity is represented by defining a set of entity and relationship paths

[0092]

[0093] in For the initial set ;|| means putting relation r into the relation path A new path is formed behind represents the relationship path propagating from the target entity i to the jth hop;

[0094] The attention weight of the target-relation path is

[0095]

[0096] The attention weight of the target-entity path is

[0097]

[0098] The final weight obtained by normalization is

[0099]

[0100] Perform a linear combination of the weight information to obtain

[0101]

[0102] Get the set of all L-hop entity embedding information

[0103] .

[0104] Optionally, the aircraft intention recognition method based on knowledge graph provided in the embodiment of the present application obtains the intention data of the target aircraft through the embedding propagation method, and the intention data includes the probability score of the target intention, specifically including:

[0105] Will gather Inside Perform a sum operation to obtain a new embedding :

[0106]

[0107] The collaborative information between targets is combined into the target embedding through the SVD++ method to obtain the embedding of target u

[0108]

[0109] Where T(u) represents the set of intentions that have interacted with target u;

[0110] right and Perform inner product operation to obtain the probability score of predicting target intention

[0111] .

[0112] An aircraft intention recognition method based on a knowledge graph provided in an embodiment of the present application adopts data fusion technology, which can integrate data from multiple sources to form a rich and multi-dimensional data set, thereby providing a basis for deeper analysis. At the same time, through semantic technology, it can enhance the semantic level of the data and achieve deeper data understanding and analysis.

[0113] Optionally, the embodiment of the present application further provides an aircraft intention recognition device based on a knowledge graph, comprising:

[0114] A processing module is used to obtain knowledge data of aircraft intentions, the knowledge data including entities, relationships, attributes and / or events, process the knowledge data, eliminate contradictions and ambiguities therein, and merge duplicate knowledge data;

[0115] A generation module is used to generate a knowledge graph of target intent based on the processed knowledge data;

[0116] The acquisition module is used to obtain the intention data of the target aircraft based on the knowledge graph of the target intention.

[0117] The present application also provides a computer-readable storage medium on which a computer program is stored, and when the program is executed by a processor, the steps of the above method are implemented. The computer-readable storage medium may include, but is not limited to, any type of disk, including a floppy disk, an optical disk, a DVD, a CD-ROM, a micro drive, and a magneto-optical disk, a ROM, a RAM, an EPROM, an EEPROM, a DRAM, a VRAM, a flash memory device, a magnetic card or an optical card, a nanosystem (including a molecular memory IC), or any type of medium or device suitable for storing instructions and / or data.

[0118] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0119] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0120] In the several embodiments provided in the present application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of the units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0121] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0122] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0123] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a memory, including a number of instructions to enable a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned memory includes: U disk, read-only memory (ROM), random access memory (RAM), mobile hard disk, disk or optical disk and other media that can store program codes.

[0124] A person skilled in the art may understand that all or part of the steps in the various methods of the above embodiments may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable memory, which may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0125] The above is only an exemplary embodiment of the present disclosure, and the scope of the present disclosure cannot be limited thereto. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the specification and practicing the disclosure here, those skilled in the art will easily think of the implementation scheme of the present disclosure. This application is intended to cover any modification, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the technical field not recorded in the present disclosure. The description and examples are regarded as exemplary only, and the scope and spirit of the present disclosure are defined by the claims.

[0126] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0127] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A method for aircraft intention recognition based on knowledge graph, characterized in that: include: Acquire knowledge data of aircraft intent, the knowledge data including entities, relationships, attributes and / or events, process the knowledge data, eliminate contradictions and ambiguities therein, and merge duplicate knowledge data; Generate a knowledge graph of target intent based on the processed knowledge data; Based on the knowledge graph of target intention, the intention data of the target aircraft is obtained; The target intention-based knowledge graph acquires the intention data of the target aircraft, including: In the knowledge graph of target intent, the target aircraft is connected to the entity through a direct relationship or an indirect relationship path; Taking the direct relationship or indirect relationship path as the attribute of the target aircraft, and acquiring the intention data of the target aircraft through an embedding propagation method; The using the direct relationship or indirect relationship path as the attribute of the target aircraft specifically includes: The propagation information of the target entity is represented by defining a set of entity and relationship paths in For the initial set ;|| means putting relation r into the relation path A new path is formed behind represents the relationship path propagating from the target entity i to the jth hop; The attention weight of the target-relation path is The attention weight of the target-entity path is The final weight obtained by normalization is Perform a linear combination of the weight information to obtain Get the set of all L-hop entity embedding information ； The intention data of the target aircraft is obtained by the embedded propagation method, and the intention data includes a probability score of the target intention, specifically including: Will gather Inside Perform a sum operation to obtain a new embedding : Combine the collaborative information between targets into the target embedding to obtain the embedding of target u Where T(u) represents the set of intentions that have interacted with target u; right and Perform inner product operation to obtain the probability score of predicting the target intention 。 2. The method for aircraft intention recognition based on knowledge graph according to claim 1, characterized in that: The knowledge graph of the target intent includes a pattern layer and a data layer; The model layer is built on the data layer and describes formulas, rules, conceptual entities and relationships in a formal way; The data layer contains the associations and relationships between instances and is the instance representation of the model layer.

3. The method for aircraft intention recognition based on knowledge graph according to claim 2, characterized in that: The model layer includes an attribute layer, a feature layer and an intention layer; The attribute layer includes target-specific attribute information, which is a parameter used for target intent analysis; The feature layer uses features as basic nodes, and the node information of the feature layer includes feature types and feature parameters; The node information of the intent layer includes the inclusion relationship, relationship value domain and relationship constraint between target entity intents.

4. An aircraft intention recognition device based on knowledge graph, characterized in that: include: A processing module, used to obtain knowledge data of aircraft intent, the knowledge data including entities, relationships, attributes and / or events, process the knowledge data, eliminate contradictions and ambiguities therein, and merge duplicate knowledge data; A generation module is used to generate a knowledge graph of target intent based on the processed knowledge data; An acquisition module, used to acquire the intention data of the target aircraft based on the knowledge graph of the target intention; The target intention-based knowledge graph acquires the intention data of the target aircraft, including: In the knowledge graph of target intent, the target aircraft is connected to the entity through a direct relationship or an indirect relationship path; Taking the direct relationship or indirect relationship path as the attribute of the target aircraft, and acquiring the intention data of the target aircraft through an embedding propagation method; The using the direct relationship or indirect relationship path as the attribute of the target aircraft specifically includes: The propagation information of the target entity is represented by defining a set of entity and relationship paths in For the initial set ;|| means putting relation r into the relation path A new path is formed behind represents the relationship path propagating from the target entity i to the jth hop; The attention weight of the target-relation path is The attention weight of the target-entity path is The final weight obtained by normalization is Perform a linear combination of the weight information to obtain Get the set of all L-hop entity embedding information ； The intention data of the target aircraft is obtained by the embedded propagation method, and the intention data includes a probability score of the target intention, specifically including: Will gather Inside Perform a sum operation to obtain a new embedding : Combine the collaborative information between targets into the target embedding to obtain the embedding of target u Where T(u) represents the set of intentions that have interacted with target u; right and Perform inner product operation to obtain the probability score of predicting the target intention 。 5. An aircraft intention recognition device based on knowledge graph, characterized in that: The method comprises at least one processing unit and at least one storage unit, wherein the storage unit stores a computer program, and when the computer program is executed by the processing unit, the processing unit executes the steps of the method according to any one of claims 1 to 3.

6. A storage medium, characterized in that: It stores a computer program that can be executed by an aircraft intention recognition device based on a knowledge graph. When the computer program runs on the aircraft intention recognition device based on a knowledge graph, the aircraft intention recognition device based on the knowledge graph executes the steps of the method described in any one of claims 1 to 3.

7. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method for aircraft intention recognition based on a knowledge graph as described in any one of claims 1 to 3 are implemented.

Images