A Method for Constructing an Air Traffic Control Information Standard System Based on Semantic Self-Clustering

Abstract

This invention provides a method for constructing an air traffic control information standard system based on semantic self-clustering. The method includes: collecting currently valid air traffic control information standards and identifying future air traffic control information standards as input for constructing the air traffic control information standard system; constructing the first layer of the air traffic control information standard system from seven aspects using an artificial tree diagram method; extracting features from the air traffic control information standard text, determining feature dimensions, constructing a feature space, and determining the third layer of the air traffic control information standard system based on similarity calculation criteria; re-extracting features from the third layer of content, determining feature dimensions, and determining the second layer of the air traffic control information standard system based on similarity calculation criteria; and finally forming the air traffic control information standard system. This invention provides technical support for the construction of the next-generation digital air traffic control information standard system.

Description

Technical Field

[0001] This invention relates to a method for constructing an air traffic control information standard system, and in particular to a method for constructing an air traffic control information standard system based on semantic self-clustering. Background Technology

[0002] Traditional standards systems built using manual tree diagrams are typically stored as documents after completion, resulting in significant underutilization and inconvenience. Furthermore, the standards system faces challenges in timely updates and responses over time. Therefore, a more dynamic and up-to-date standards system is needed. Traditional air traffic control information standards management platforms rely on post-event input from participants, leading to issues with the timeliness and accuracy of technical information and increasing the risk of information security incidents due to the large number of participants. Additionally, most air traffic control standards are stored across different participating units, hindering timely and comprehensive recording of relevant information and content, thus complicating standard review and use. Traditional modeling methods like manual tree diagrams also result in weak dynamism and scalability in air traffic control information standards systems. Summary of the Invention

[0003] Purpose of the invention: The technical problem to be solved by the present invention is to provide a method for constructing an air traffic control information standard system based on semantic self-clustering, which addresses the shortcomings of the existing technology.

[0004] To address the aforementioned technical problems, this invention discloses a method for constructing an air traffic control information standard system based on semantic self-clustering, comprising the following steps:

[0005] Step 1: Collect currently valid air traffic control information standard data to form text information, which serves as the information input for constructing an air traffic control information standard system based on semantic self-clustering; the currently valid air traffic control information standard data includes: the name of the air traffic control information standard, the number of the air traffic control information standard, the category of the air traffic control information standard, the status of the air traffic control information standard, the content of the air traffic control information standard, and the attributes of the air traffic control information standard.

[0006] Step 2: Determine future air traffic control information standards, including the standard's name, scope of application, specific content, and relevant business area, as information input for constructing an air traffic control information standard system based on semantic self-clustering;

[0007] Step 3: Based on the artificial tree diagram method, construct the first layer of the air traffic control information standard system from seven aspects, including: basic general information, airspace information, air traffic flow management information, air traffic services, communication, navigation and surveillance, aviation meteorology, and aeronautical information;

[0008] The method based on artificial tree diagrams includes the following steps:

[0009] Step 3-1 uses the structural decomposition method to decompose the content of the first layer of the air traffic control information standard system. The first layer fully includes the current effective air traffic control information standards and future air traffic control information standards. The users in the air traffic control information standard system include: military aviation, civil aviation, general aviation, and UAV users. The dimensions in the air traffic control information standard system include: communication, navigation, surveillance, meteorology, control, traffic flow, and airspace dimensions. The attributes in the air traffic control information standard system include: common infrastructure, common support, and business applications.

[0010] Step 3-2: Number the first layer of content in the air traffic control information standard system using a combination of numbers and letters.

[0011] Step 4: Based on the text information in Step 1 and the air traffic control information standards in Step 2, perform feature extraction, determine the feature dimensions, and form the air traffic control information standard feature space, including the following steps:

[0012] 4-1: Read the text information mentioned in step 1 and the content corresponding to the standard name, standard introduction and standard scope in the future air traffic control information standard mentioned in step 2;

[0013] 4-2: Use the cut method in jieba technology to segment the text information read in 4-1 to obtain word terms;

[0014] 4-3: Use regular expressions to clean up useless tags, special symbols, and stop words;

[0015] 4-4: Calculate the word frequency of the terms after word segmentation in step 4-2, and sort the original features, i.e., the terms, according to the statistical results in descending order of word frequency;

[0016] 4-5: Based on the statistical results of step 4-4, feature extraction is performed on the text information contained in each standard using the TF-IDF method. The specific method is as follows:

[0017]

[0018] Among them, |D i | represents the total number of terms contained in the i-th standard Chinese text information, i∈(1,2,3,…N), |D l | represents the total number of terms l in the i-th standard text, N represents the total number of currently valid air traffic control information standard data and future air traffic control information standards, and l∈(1,2,3,…20) represents the total number of feature values ​​in the document corresponding to the i-th standard, determined by experts; idf iThe term proportion is used to select words representing each standard topic as features, ranked from smallest to largest. These selected words are then sorted and stored as a feature list, denoted as the key. i Repeat the above steps for each standard to complete the feature extraction.

[0019] The determination of feature dimensions refers to determining the total number l of feature values ​​mentioned in steps 4-5; the feature dimensions are determined based on expert advice.

[0020] The aforementioned formation of the air traffic control information standard feature space refers to combining the current valid air traffic control information standard data with the feature list (key) of future air traffic control information standards. i Consider it as a point in space; all points form the standard feature space of air traffic control information.

[0021] Step 5: Under the air traffic control information standard feature space, based on the similarity calculation criterion, different standards are combined into k clusters, and the names of the clusters are determined; the names of the determined k clusters are used as the third layer of the air traffic control information standard system.

[0022] The aforementioned similarity calculation criterion refers to the calculation method based on Jaccard similarity, which includes the following steps:

[0023] Step 5-1: Calculate the similarity of all keysi, using the following method:

[0024]

[0025] Where i ≠ i; key i Represents the feature list of the i-th criterion; key j This represents the feature list of the j-th criterion;

[0026] Step 5-2: Calculate the Jaccard distance dis based on this distance. The calculation method is as follows:

[0027]

[0028] Step 5-3: Set the Jaccard distance (dis) threshold to 0. if If the Jaccard distance is less than or equal to the threshold, then the key is determined. i and key j resemblance.

[0029] The determination of the names of the k clusters as the third layer of the air traffic control information standard system refers to the information content of the range and business attributes of the standards corresponding to the names of the k clusters determined in step 5.

[0030] The value of k is a hyperparameter, specified by experts; the names of the k clusters are determined by experts based on the air traffic control information standard attributes.

[0031] Step 6: Re-extract features from the third layer content, determine the feature dimensions, form a new feature space, and reclassify the third layer content based on the similarity calculation criteria to form m new clusters. Determine the names of the new clusters as the second layer of the air traffic control information standard system.

[0032] The value of m is a hyperparameter, specified by experts, and the names of the m new clusters are determined by experts.

[0033] Step 7: Repeat the above steps, combining bottom-up and top-down approaches, to finally form an air traffic control information standard system based on semantic self-clustering, thus completing the construction of the air traffic control information standard system based on semantic self-clustering.

[0034] Beneficial effects:

[0035] This invention provides a method for constructing an air traffic control information standard system based on semantic self-clustering. From the perspective of facilitating timely use and updating of air traffic control information standards by air traffic control practitioners, it assists air traffic control personnel in flexibly using, constructing, and updating the air traffic control information standard system. This solves the problems of weak dynamism and scalability of air traffic control information standard systems constructed by traditional manual tree diagram and other modeling methods, and provides technical support for the construction of the next generation of digital air traffic control information standard systems. Attached Figure Description

[0036] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments, and the advantages of the present invention in the above and / or other aspects will become clearer.

[0037] Figure 1 This is a schematic diagram of the workflow of the present invention.

[0038] Figure 2 This is a schematic diagram of the design of an air traffic control information standard system that combines bottom-up and top-down approaches.

[0039] Figure 3 This is a schematic diagram of the self-clustering algorithm in this invention. Detailed Implementation

[0040] like Figure 1As shown, this invention discloses a method and tool for constructing an air traffic control information standard system based on semantic self-clustering, belonging to the field of air traffic management standard systems. The method first collects currently valid air traffic control information standards and the air traffic control information standards required by the air traffic control industry for the next 10 years as input for constructing the air traffic control information standard system. Then, based on the artificial tree diagram method, it constructs the first layer of the national air traffic control information standard system framework from seven aspects: basic general information, airspace information, air traffic flow management information, air traffic services, communication, navigation, and monitoring, aviation meteorology, and aviation intelligence. Next, it extracts features from the air traffic control information standard data, determines feature dimensions, and forms an air traffic control information standard feature space. Under the air traffic control information standard feature space, based on similarity calculation criteria, multiple standards are combined into k clusters, and the names of the determined k clusters serve as the third layer of the national air traffic control information standard system. Finally, the information corresponding to the k clusters formed in the third layer is reclassified to form m new clusters, which serve as the second layer of the national air traffic control information standard system. Repeating the above steps, through a combination of bottom-up and top-down approaches, a reasonable national air traffic control information standard system is finally formed.

[0041] This invention primarily addresses the issue of weak dynamism and scalability in constructing air traffic control information standards systems. A rapid construction system for air traffic control information standards systems based on this method overcomes the limitations of traditional manual tree diagram and other modeling methods in terms of dynamism and scalability. Secondly, this invention provides technical support and reference for the construction of digital systems for standards systems. The specific steps are as follows:

[0042] Step 1: Collect currently valid air traffic control information standard data to form text information, including the name, number, category, status, content, and attributes of the air traffic control information standard. This serves as the information input for constructing an air traffic control information standard system based on semantic self-clustering.

[0043] Step 2: Through research and analysis, and discussions with industry experts, determine the air traffic control information standards required by the air traffic control industry in the next 10 years, including the name of the standard, its scope of application, the specific content of the standard, and the business area to which it belongs, as the information input for the construction of an air traffic control information standard system based on semantic self-clustering.

[0044] Step 3: Taking full account and drawing upon the relevant standards classification principles of the International Civil Aviation Organization (ICAO), the existing air traffic control standards systems of both military and civil aviation were compiled. Opinions and suggestions were solicited from air traffic control equipment manufacturers, research institutions, and other air traffic control experts. Based on the structural decomposition method, the first layer of the air traffic control information standards system was decomposed. This first layer must fully encompass both currently effective air traffic control information standards and future air traffic control information standards. The first layer includes seven aspects: basic general information, airspace information, air traffic flow management information, air traffic services, communication, navigation, and surveillance (CNS), aviation meteorology, and aeronautical information. Figure 2 As shown; the users in the air traffic control information standard system include: military aviation, civil aviation, general aviation and UAV users; the dimensions in the air traffic control information standard system include: communication, navigation, surveillance, meteorology, control, traffic flow and airspace dimensions; the attributes in the air traffic control information standard system include: common foundation, common support and business application; and the first layer of the air traffic control information standard system is numbered in a combination of numbers and letters.

[0045] Step 4: Based on Step 3, based on... Figure 3 The self-clustering method shown extracts features from the existing valid air traffic control information standard data formed in steps 1 and 2, as well as the air traffic control information standard data required for the air traffic control industry standards in the next 10 years. The steps include:

[0046] 4-1: Read the text information mentioned in step 1 and the content corresponding to the standard name, standard introduction and standard scope in the future air traffic control information standard mentioned in step 2;

[0047] 4-2: Use the cut method in jieba technology to segment the text information read in 4-1 to obtain word terms; the essence of the cut method is to decompose the text information into word terms according to certain rules through a for loop;

[0048] 4-3: Use regular expressions to clean up useless tags, special symbols, and stop words; regular expressions describe a string matching pattern, which can be used to check whether a string contains a certain substring and extract the substring that meets a certain condition from a string.

[0049] 4-4: Calculate the word frequency of the terms after word segmentation in step 4-2, and sort the original features, i.e., the terms, according to the statistical results in descending order of word frequency;

[0050] 4-5: Based on the statistical results of step 4-4, feature extraction is performed on the text information contained in each standard using the TF-IDF method. The specific method is as follows:

[0051]

[0052] Among them, |Di | represents the total number of terms contained in the i-th standard text information, i∈(1,2,3,…N), |D l | represents the total number of terms containing the i-th standard text, N represents the total number of currently valid air traffic control information standard data and future air traffic control information standards, and l∈(1,2,3,…20) represents the total number of feature values ​​contained in the document corresponding to the i-th standard, which is determined by experts and is 20 in this invention; idf i The term proportion is used to select words representing each standard topic as features, ranked from smallest to largest. These selected words are then sorted and stored as a feature list, denoted as the key. i Repeat the above steps for each standard to complete the feature extraction.

[0053] Based on feature extraction, the number of standard information features to be extracted in step 4-3 is determined; in this invention, based on expert advice, the feature dimension is determined to be 20.

[0054] Finally, forming the air traffic control information standard feature space refers to compiling a list of key features of currently valid air traffic control information standard data and the air traffic control information standards required for the air traffic control industry standards of the next 10 years. i Consider it as a point in space; all such points form the feature space of air traffic control information standards.

[0055] Step 5: Under the standard feature space of air traffic control information formed in Step 4, the similarity calculation method based on Jaccard similarity includes the following steps:

[0056] Step 5-1: Calculate the similarity of all keysi using the formula shown below.

[0057]

[0058] In the formula, i ≠ j;

[0059] Step 5-2: Calculate the Jaccard distance based on this distance. The calculation formula is as follows:

[0060]

[0061] Step 5-3: Set the Jaccard distance threshold if If it is less than the threshold, it means the key... i and key j resemblance.

[0062] Multiple standards are combined into k clusters, where the value of k is a hyperparameter specified by experts; the names of the k clusters are determined by experts based on the attributes of the air traffic control information standards; the determined names of the k clusters serve as the third layer of the national air traffic control information standard system.

[0063] Step 6: Re-extract features from the third layer of the national air traffic control information standard system containing k clusters, including the names of the k clusters named by experts in Step 5, as well as the scope of standards and business attributes corresponding to each cluster name, determine the feature dimensions, and form a new feature space. Based on the similarity calculation criteria in Step 5, reclassify the third layer of the national air traffic control information standard system to form m new clusters. The value of m is a hyperparameter, specified by experts. The names of the m new clusters are determined by experts and serve as the second layer of the national air traffic control information standard system.

[0064] Step 7: Repeat the above steps, and through a combination of bottom-up and top-down approaches, finally form a reasonable national air traffic control information standard system.

[0065] In its specific implementation, this application provides a computer storage medium and a corresponding data processing unit. The computer storage medium is capable of storing a computer program, which, when executed by the data processing unit, can run the invention's content regarding a method for constructing an air traffic control information standard system based on semantic self-clustering, as well as some or all of the steps in various embodiments. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0066] Those skilled in the art will clearly understand that the technical solutions in the embodiments of the present invention can be implemented using computer programs and their corresponding general-purpose hardware platforms. Based on this understanding, the technical solutions in the embodiments of the present invention, or the parts that contribute to the prior art, can be embodied in the form of computer programs, i.e., software products. These computer program software products can be stored in a storage medium and include several instructions to cause a device containing a data processing unit (which may be a personal computer, server, microcontroller, MUU, or network device, etc.) to execute the methods described in various embodiments or certain parts of the embodiments of the present invention.

[0067] This invention provides an idea and method for constructing an air traffic control information standard system based on semantic self-clustering. Many methods and approaches exist for implementing this technical solution; the above description is merely a preferred embodiment of the invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of this invention, and these improvements and modifications should also be considered within the scope of protection of this invention. All components not explicitly stated in this embodiment can be implemented using existing technologies.

Claims

1. A method for constructing an air traffic control information standard system based on semantic self-clustering, characterized in that, Includes the following steps: Step 1: Collect existing and valid air traffic control information standard data to form text information, which serves as the information input for constructing an air traffic control information standard system based on semantic self-clustering; Step 2: Determine future air traffic control information standards, including the standard's name, scope of application, specific content, and relevant business area, as information input for constructing an air traffic control information standard system based on semantic self-clustering; Step 3: Based on the artificial tree diagram method, construct the first layer of the air traffic control information standard system from seven aspects, including: basic general information, airspace information, air traffic flow management information, air traffic services, communication, navigation and surveillance, aviation meteorology, and aeronautical information; Step 4: Extract features from the text information in Step 1 and the air traffic control information standard in Step 2, determine the feature dimensions, and form the air traffic control information standard feature space. Step 5: Under the air traffic control information standard feature space, based on the similarity calculation criterion, different standards are combined into k clusters, and the names of the clusters are determined; the names of the determined k clusters are used as the third layer of the air traffic control information standard system. Step 6: Re-extract features from the third layer content, determine the feature dimensions, form a new feature space, and reclassify the third layer content based on the similarity calculation criteria to form m new clusters. Determine the names of the new clusters as the second layer of the air traffic control information standard system. Step 7: Repeat the above steps, combining bottom-up and top-down approaches, to finally form an air traffic control information standard system based on semantic self-clustering, thus completing the construction of the air traffic control information standard system based on semantic self-clustering.

2. The method for constructing an air traffic control information standard system based on semantic self-clustering according to claim 1, characterized in that, The currently valid air traffic control information standard data mentioned in step 1 includes: the name of the air traffic control information standard, the number of the air traffic control information standard, the category of the air traffic control information standard, the status of the air traffic control information standard, the content of the air traffic control information standard, and the attributes of the air traffic control information standard.

3. The method for constructing an air traffic control information standard system based on semantic self-clustering according to claim 2, characterized in that, The method based on artificial tree diagrams described in step 3 includes the following steps: Step 3-1 uses the structural decomposition method to decompose the content of the first layer of the air traffic control information standard system. The first layer fully includes the current effective air traffic control information standards and future air traffic control information standards. The users in the air traffic control information standard system include: military aviation, civil aviation, general aviation, and UAV users. The dimensions in the air traffic control information standard system include: communication, navigation, surveillance, meteorology, control, traffic flow, and airspace dimensions. The attributes in the air traffic control information standard system include: common infrastructure, common support, and business applications. Step 3-2: Number the first layer of content in the air traffic control information standard system using a combination of numbers and letters.

4. The method for constructing an air traffic control information standard system based on semantic self-clustering according to claim 3, characterized in that, The feature extraction described in step 4 includes the following steps: 4-1: Read the text information mentioned in step 1 and the content corresponding to the standard name, standard introduction and standard scope in the future air traffic control information standard mentioned in step 2; 4-2: Use the cut method in jieba technology to segment the text information read in 4-1 to obtain word terms; 4-3: Use regular expressions to clean up useless tags, special symbols, and stop words; 4-4: Calculate the word frequency of the terms after word segmentation in step 4-2, and sort the original features, i.e., the terms, according to the statistical results in descending order of word frequency; 4-5: Based on the statistical results of step 4-4, feature extraction is performed on the text information contained in each standard using the TF-IDF method. The specific method is as follows: Among them, |D i | represents the total number of terms contained in the i-th standard Chinese text information, i∈(1,2,3,…N), |D l | represents the total number of terms l in the i-th standard text, N represents the total number of currently valid air traffic control information standard data and future air traffic control information standards, and l∈(1,2,3,…20) represents the total number of feature values ​​in the document corresponding to the i-th standard, determined by experts; idf i The term proportion is used to select words representing each standard topic as features, ranked from smallest to largest. These selected words are then sorted and stored as a feature list, denoted as the key. i Repeat the above steps for each standard to complete the feature extraction.

5. The method for constructing an air traffic control information standard system based on semantic self-clustering according to claim 4, characterized in that, The determination of feature dimensions mentioned in step 4 refers to determining the total number l of feature values ​​mentioned in steps 4-5; the feature dimensions are determined based on expert advice.

6. The method for constructing an air traffic control information standard system based on semantic self-clustering according to claim 5, characterized in that, The formation of the air traffic control information standard feature space mentioned in step 4 refers to combining the current valid air traffic control information standard data with the feature list keys of future air traffic control information standards. i Consider it as a point in space; all points form the standard feature space of air traffic control information.

7. The method for constructing an air traffic control information standard system based on semantic self-clustering according to claim 6, characterized in that, The similarity calculation criterion mentioned in step 5 refers to the calculation method based on Jaccard similarity, which includes the following steps: Step 5-1: Calculate the similarity of all keysi, using the following method: Where i≠j; key i Represents the feature list of the i-th criterion; key j This represents the feature list of the j-th criterion; Step 5-2: Calculate the Jaccard distance dis based on this distance. The calculation method is as follows: Step 5-3: Set the Jaccard distance (dis) threshold to 0. if If the Jaccard distance is less than or equal to the threshold, then the key is determined. i and key j resemblance.

8. The method for constructing an air traffic control information standard system based on semantic self-clustering according to claim 7, characterized in that, The determination of the names of the k clusters in step 5 as the third layer of the air traffic control information standard system refers to the information content of the range and business attributes of the standards corresponding to the names of the k clusters determined in step 5.

9. The method for constructing an air traffic control information standard system based on semantic self-clustering according to claim 8, characterized in that, The value of k mentioned in step 5 is a hyperparameter, specified by experts; the names of the k clusters are determined by experts based on the air traffic control information standard attributes.

10. The method for constructing an air traffic control information standard system based on semantic self-clustering according to claim 9, characterized in that, The value of m mentioned in step 6 is a hyperparameter, specified by experts, and the names of the m new clusters are determined by experts.

Images