KR20210086402A - Apparatus and methods for trend analysis in airport and aviation technology - Google Patents

Abstract

The present invention relates to an apparatus and a method for analyzing the trend of airport and aviation technologies, which filter data on the airport and aviation technologies from data collected from web content, extract keywords by analyzing morphemes from the filtered data, and analyze trends by analyzing the association between the extracted keywords. In accordance with an embodiment of the present invention, a method for analyzing the trend of airport and aviation technologies comprises the steps of: collecting documents related to the airport and aviation technologies from a web; extracting text from the collected documents and identifying morphemes from the extracted text; extracting a noun from among the identified morphemes; generating a document word matrix using the extracted nouns; and analyzing the degree of similarity and relevance between the keywords corresponding to each element of the document word matrix and the degree of similarity and relevance between the respective documents using the document word matrix, and providing an analysis result.

Description

Apparatus and methods for trend analysis in airport and aviation technology}

The present invention relates to an apparatus and method for analyzing trends in airport and aviation technology, and more particularly, filtering data on airport and aviation technology from data collected from web content, extracting keywords by analyzing morphemes from the filtered data, and , to a trend analysis device and method for airport and aviation technology that analyzes trends by analyzing the correlation between extracted keywords.

Recently, a keyword analysis system that analyzes a large amount of freely described text data and obtains useful information from the analysis result is attracting attention. The keyword analysis system is a technique that analyzes after converting into structured data to obtain useful information from unstructured data made of natural language. A keyword analysis system may be applied to an Internet search engine or a search system in a reading room, and documents may be classified according to an existing given category.

However, conventional keyword analysis systems do not specialize in aviation data and can only grasp trends in SNS information or news, and there is a limit to not knowing detailed information about technology, and information such as patents, tasks, and academics is separately available for each site. There is a problem in that the information of airports and aviation technology is fragmented due to operation.

In addition, the core of keyword analysis technology is to extract core nouns through analysis of morphemes of characters. Existing technologies have a problem in that they cannot be recognized as nouns without information on core nouns. Accordingly, there is a need for a method for extracting new nouns or compound nouns for new aviation technology.

In addition, according to the 4th industrial revolution, airport and aviation industry technologies are rapidly changing and developing, and accordingly, it is necessary to analyze and predict the technology trends of the airport and aviation industry to lead the technology industry through technology development that fits the flow of the industry. There is a need for a solution for this.

The present invention is to solve the problems described above, by filtering data on airport and aviation technology from data collected from web content, extracting keywords by analyzing morphemes from the filtered data, and analyzing the correlation between the extracted keywords. The purpose of this is to provide an apparatus and method for analyzing trends in airport and aviation technology to analyze trends.

In addition to the technical problems of the present invention mentioned above, other features and advantages of the present invention will be described below or will be clearly understood by those skilled in the art from such description and description.

The trend analysis method of airport and aviation technology according to an embodiment of the present invention for achieving the above-described object includes the steps of collecting documents related to airport and aviation technology from the web, extracting text from the collected documents, and extracting the Identifying morphemes from text, extracting nouns from among the identified morphemes, generating a document word matrix using the extracted nouns, and using the document word matrix to correspond to each element of the document word matrix and analyzing the degree of similarity and relevance between keywords, and the degree of similarity and relevance between the respective documents, and providing an analysis result.

On the other hand, the airport and aviation technology trend analysis apparatus according to an embodiment of the present invention for achieving the above-described object includes a memory for storing documents collected in relation to airport and aviation technology from the web, and texts from the collected documents. extracting, identifying morphemes from the extracted text, extracting nouns from among the identified morphemes, generating a document word matrix using the extracted nouns, and using the document word matrix to correspond to each element of the document word matrix. The processor may include a processor that analyzes the degree of similarity and relevance between keywords and the degree of similarity and relevance between each document, and provides an analysis result.

An apparatus and method for analyzing airport and aviation technology trends according to an embodiment of the present invention filters data on airport and aviation technology from data collected from web content, and extracts keywords by analyzing morphemes from the filtered data. Trends can be analyzed by analyzing the correlation between keywords.

In addition, it is possible to collect and integrate information on airports and aviation technologies that are distributed to several existing sites, solve the problem of fragmentation of information, and enable unified information analysis.

A more accurate and objective technical analysis may be possible through an unstructured analysis of airport and aviation technology through a number of unstructured algorithms and an analysis that combines a number of analysis results according to the algorithm.

In addition, by extracting key keywords of technology through the integration of airport and aviation technology information and analyzing the amount of change in the number of exposures of keywords, the direction of development of airport and aviation technology over time is identified, and technology development in the rapidly changing technology market can be predicted.

In addition, other features and advantages of the present invention may be newly recognized through embodiments of the present invention.

1 is a system for analyzing trends in airport and aviation technology according to an embodiment of the present invention.
2 is a diagram illustrating an apparatus for analyzing trends in airport and aviation technology according to an embodiment of the present invention.
3 is a diagram illustrating extracting text from collected documents according to an embodiment of the present invention.
4 is a diagram illustrating analysis of morphemes and extraction of nouns from the extracted text according to an embodiment of the present invention.
5 is a diagram illustrating a document word matrix according to an embodiment of the present invention.
6 is a diagram illustrating a keyword map according to an embodiment of the present invention.
7 is a diagram illustrating sorting of keywords according to an embodiment of the present invention.
8 is a diagram illustrating sorting of sets according to an embodiment of the present invention.
9 is a diagram illustrating a method for analyzing trends in airport and aviation technology according to an embodiment of the present invention.

In order to clearly explain the present invention, parts irrelevant to the description are omitted, and the same reference numerals are given to the same or similar elements throughout the specification.

Although not defined otherwise, all terms including technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the art to which the present invention belongs. Commonly used terms defined in the dictionary are additionally interpreted as having a meaning consistent with the related technical literature and the presently disclosed content, and unless defined, they are not interpreted in an ideal or very formal meaning.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those of ordinary skill in the art can easily implement them. However, the present invention may be embodied in various different forms and is not limited to the embodiments described herein.

1 is a system for analyzing trends in airport and aviation technology according to an embodiment of the present invention.

Referring to FIG. 1 , the airport and aviation technology trend analysis system 1000 according to an embodiment of the present invention may include a web 100 and an airport and aviation technology trend analysis apparatus 200 (hereinafter, a trend analysis apparatus). have.

The web 100 may generally refer to an Internet space. Information such as news, academics, R&D projects, patents and policies may be disclosed on the web 100 , and each information may be disclosed in each site where the corresponding information is disclosed. In this case, each site may be an HTML (Hyper Text Markup Language) document. For example, news information may be published on a news site, and the news site may be an HTML document.

The trend analysis device 200 may collect information such as news, academics, R&D projects, patents and policies from the web 100 from each site. In this case, the trend analysis apparatus 200 may collect HTML documents from each site. Here, the trend analysis apparatus 200 may collect documents about airports and aviation technologies from the web 100 . Also, the trend analysis apparatus 200 may filter documents about airports and aviation technologies among documents collected from the web 100 , and store the filtered documents about airports and aviation technologies.

The trend analysis apparatus 200 may extract text from documents (or filtered documents) collected from each site, and identify morphemes from the extracted text. The trend analysis apparatus 200 may extract nouns from the identified morphemes and convert the extracted nouns into a document word matrix. The trend analysis apparatus 200 may analyze the degree of similarity and relevance between documents collected from each site by using the converted document word matrix, and may provide the analyzed result to the user. Also, the trend analysis apparatus 220 may analyze the degree of similarity and relevance between the extracted nouns using the converted document word matrix, and may provide the analyzed result to the user. Here, the trend analysis device 200 may display the analyzed result on the display through the output device. In this case, the output device may be one configuration included in the trend analysis device 200 or a separate configuration.

The trend analysis apparatus 200 may analyze a trend of information disclosed in web content by extracting and analyzing nouns from documents collected from the web 100 . For example, in the case of a noun that is described with a lot of frequency in documents of various fields, it may be a noun that is used a lot recently. Accordingly, a recent trend may be analyzed by analyzing nouns collected from the web 100 on the basis of nouns that have been frequently used recently.

2 is a diagram illustrating an apparatus for analyzing trends in airport and aviation technology according to an embodiment of the present invention.

Referring to FIG. 2 , an apparatus 200 for analyzing trends in airport and aviation technology according to an embodiment of the present invention (hereinafter, a trend analysis apparatus) may include a memory 210 and a processor 220 .

The memory 210 may store information collected from the web 100 . That is, news information, academic information, R&D project information, patent information, policy information, and the like collected from the web 100 may be stored. Here, the information collected from the web 100 may be an HTML document of a news, academic, R&D project, patent, or policy site.

Also, a morpheme dictionary may be stored in the memory 210 . The morpheme dictionary may store morphemes. The morpheme dictionary may be configured in advance and stored in the memory 210 , may be updated by information collected from the web 100 , or may be updated by a user input. Here, the morphemes stored in the morpheme dictionary may be morphemes identified from words related to airports and aviation.

The processor 220 may collect representative keywords of each document by analyzing the HTML tag of the site where the documents collected from the web 100 are disclosed. HTML tags can represent structural semantics for the body, such as headings, paragraphs, lists, etc. Accordingly, by analyzing the HTML tag, the body, that is, the title, paragraph, list, etc. of the document disclosed on the site can be determined. The processor 220 may collect representative keywords that are mainly disclosed in the corresponding document through the determined title, paragraph, list, and the like.

For example, the processor 220 may analyze an HTML tag of a site where a patent document related to 'boarding' is disclosed. The processor 220 may analyze the HTML tag to determine the title, paragraph, list, etc. of the document, and as a result of the determination, determine that the representative keyword mainly disclosed in the document is 'boarding', and collect them. have.

The processor 220 may update the morpheme dictionary based on the collected representative keywords. The processor 220 may update the morpheme dictionary based on keywords not stored in the morpheme dictionary among the collected representative keywords. Here, the processor 220 may convert the representative keywords into a morpheme form, and update the representative keyword converted into the morpheme form in advance. In this case, the morpheme updated in the morpheme dictionary may be a morpheme converted from keywords related to airports and airlines. That is, keywords related to airport and aviation technology may be stored in a morpheme dictionary in a morpheme form.

The processor 220 may extract text from documents collected from the web 100 and identify morphemes from the extracted text. The processor 220 may extract text from documents collected through text mining. The processor 220 may identify the morpheme from the extracted text by comparing the extracted text with the morphemes stored in the morpheme dictionary, respectively. Here, the processor 220 may extract a text for each of the collected documents, and may identify a morpheme from each of the extracted texts. The processor 220 may identify the same text as the morpheme stored in the morpheme dictionary from each extracted text as a morpheme.

The processor 220 may extract a noun by deleting a stopword from among the identified morphemes, and converting a synonym from among the identified morphemes into a single morpheme. Here, the stopwords may include words, articles, prepositions, articles, conjunctions, etc. that are not used as terms. That is, since an article, a preposition, a proposition, a conjunction, and the like are deleted from the morpheme identified by the processor 220 , only the morpheme in the form of a noun may remain. Accordingly, the processor 220 may extract a noun from among the identified morphemes. Also, in the case of synonyms such as 'age' and 'age', the processor 220 may convert morphemes to be unified into one of 'age' or 'age'. By converting the synonym into one morpheme by the processor 220 to unify the synonym, the morpheme (keyword) can be analyzed more accurately. Here, the processor 220 may delete a stopword and convert a synonym from the extracted noun into a single morpheme, convert a synonym among the identified morphemes into a single morpheme, and then delete the stopword to extract the noun. have.

The processor 220 may generate a document word matrix using the extracted nouns. The document word matrix may be a matrix expressing the number of respective nouns disclosed in a plurality of documents. That is, the processor 220 may generate a document word matrix based on the number of each of the extracted nouns disclosed in each of the collected documents.

A row of the document word matrix may be the number of each extracted nouns appearing in a specific document, and a column may be the number of extracted nouns starting in each of the documents. In addition, the row of the document word matrix may be the number of extracted nouns that are disclosed in each document among the extracted nouns, and the column may be the number of extracted nouns that are disclosed in the specific document.

In addition, the number of rows of the document word matrix may be the number of extracted nouns, and the number of columns may be the number of collected documents. In addition, the number of rows of the document word matrix may be the number of collected documents, and the number of columns may be the number of extracted nouns.

The processor 220 may analyze the degree of similarity and relevance between keywords corresponding to each element of the document word matrix and the degree of similarity and relevance between documents using the document word matrix, and may provide the analyzed result to the user.

The processor 220 may calculate a distance between keywords corresponding to each element of the document word matrix, and generate a keyword map based on the calculated distance. The processor 220 may calculate the distance between the keywords and connect the keywords in a tree form.

The processor 220 may calculate a distance between keywords, classify keywords having a distance from a specific keyword within a predetermined distance for each layer, and generate a keyword map by connecting the keywords for each layer. For example, as a result of calculating the distance between keywords, keywords having a distance difference of N with respect to a specific keyword may be classified into the same layer. Also, keywords having a distance difference of M with respect to a specific keyword may be classified into the same layer. In this case, when N is a number smaller than M, a hierarchy of keywords having a distance difference of N may be located closer to a specific keyword than a hierarchy of keywords having a distance difference of M.

In addition, the processor 220 calculates the distance between the keywords to connect the keywords having a distance from the specific keyword within a predetermined distance to the specific keyword, and connects keywords having a distance from each of the keywords within a predetermined distance to the keyword within a predetermined distance. can connect with That is, the processor 220 may generate a keyword map by connecting keywords having a close distance to the specific keyword with the specific keyword, and connecting keywords having a close distance to each of the keywords connected to the specific keyword. For example, as a result of calculating the distance between the keywords, keywords 'B' and 'C' having a distance difference within N with respect to the specific keyword 'A' may be connected to the specific keyword 'A'. In addition, for each of the keywords 'B' and 'C' having a distance difference within N connected to the specific keyword, the keywords having a distance difference within M and each of the keywords having a distance difference within N connected to the specific keyword can connect That is, the keywords 'D' and 'E' having a distance difference within M from the keyword 'B' may be connected to the keyword 'B'. Also, the keyword 'F' having a distance difference within M from the key word 'C' may be connected to the keyword 'C'. Here, N and M may have the same value or different values.

The processor 220 may provide a keyword map generated according to a user's manipulation. That is, the processor 220 may display the keyword map on the output device. Accordingly, the user may easily identify related keywords in the plurality of documents.

The processor 220 may sequentially sort the keywords from the largest number of keywords to the smallest number of keywords in the document word matrix. The processor 220 may sort by changing only the order of rows or columns while maintaining the form of the document word matrix. For example, if the row of the document word matrix is the number of extracted nouns each being disclosed in documents, the rows are sequentially the number of keyword 1 disclosed in documents, the number of keyword 2 disclosed in documents, ??, The number of keywords n may be disclosed in documents. In this case, when the number of keyword 3 disclosed in documents is greater than the number of keyword 1 disclosed in documents, the processor 220 determines that the number of keyword 3 disclosed in documents is greater than the number of keyword 1 disclosed in documents. You can sort the order of keywords so that they start earlier. Here, the preceding order may mean elements having a small number of i and j of the _{matrix H ij .}

Also, the processor 220 may separately sort the keywords in an order of increasing the number of keywords. For example, the processor 220 may sort keywords in the form of (keyword 1, keyword 2, keyword 3, ??).

The processor 220 may provide sorted keywords according to a user's manipulation. That is, the processor 220 may display the sorted keywords on the output device. Accordingly, the user may easily identify main keywords in the plurality of documents.

The processor 220 may configure a set of keywords corresponding to each element of the document word matrix for each similar keyword, calculate a distance between the configured sets, and arrange the configured sets according to the distance.

The processor 220 may provide sorted sets according to a user's manipulation. That is, the processor 220 may display the sorted sets on the output device. Accordingly, the user may easily identify similar keywords in a plurality of documents.

3 is a diagram illustrating extracting text from collected documents according to an embodiment of the present invention.

Referring to FIG. 3 , the processor 220 may collect documents from the web 100 . The processor 220 may collect documents disclosed in news, academic, R&D projects, patents, and policy sites. For example, the processor 220 may collect a first document from a patent site, collect a second document from a news site, and collect a third document from a policy site. Here, the first document, the second document, and the third document collected from each site may be documents related to airport and aviation technology.

In addition, the processor 220 may filter documents about airport and aviation technology among documents collected from each site, and the result of filtering the documents about airport and aviation technology is the first document, the second document and It may be a third document. For example, the processor 220 may collect documents using the similarity between the address of the site, the name of the site, the name of the document, the content of the document, the author of the document, the host of the site, and words related to airport and aviation technology, respectively. You can filter documents about airports and aviation technologies.

The processor 220 may extract text from each document through text mining. That is, the processor 220 may extract text from the first document, extract text from the second document, and extract text from the third document.

4 is a diagram illustrating analysis of morphemes and extraction of nouns from the extracted text according to an embodiment of the present invention.

Referring to FIG. 4 , the processor 220 may compare the text extracted from each document with the morpheme dictionary stored in the memory 210 . The processor 220 may compare texts with a plurality of morphemes stored in the morpheme dictionary. The processor 220 may identify the same text as morphemes stored in the morpheme dictionary among texts as a morpheme. Here, the morphemes stored in the morpheme dictionary may be morphemes related to airport and aviation technology, and accordingly, the processor 220 may identify morphemes related to airport and aviation technology from each document. That is, the processor 220 may identify the morphemes in the documents by comparing the texts extracted from the documents with the morphemes stored in the morpheme dictionary. In this case, since the morpheme stored in the morpheme dictionary is a morpheme related to airports and aviation, morphemes identified in documents may also be morphemes related to airports and aviation.

The processor 220 may extract a noun by deleting a stopword from among the identified morphemes, and converting a synonym from among the identified morphemes into a single morpheme. Here, the stopwords may include words, articles, prepositions, articles, conjunctions, etc. that are not used as terms. That is, since an article, a preposition, a proposition, a conjunction, and the like are deleted from the morpheme identified by the processor 220 , only the morpheme in the form of a noun may remain. Accordingly, the processor 220 may extract a noun from among the identified morphemes. Also, the processor 220 may unify the synonyms by converting the synonyms into one morpheme. For example, keywords of family and family may all exist in the documents, and family and family may be synonyms used in the same meaning. In this case, by the processor 220 , the keywords of the family and the family may be unified into one keyword of the family or the family. Meanwhile, when synonyms are not unified into one keyword by the processor 220 , each keyword may be analyzed as a separate keyword. In this case, the analysis of keywords may not be accurate. For example, when family and family keywords are converted into family and analyzed, the family keyword and family keyword may be analyzed as the combined number. Meanwhile, when family and family keywords are analyzed as separate keywords, each number may be analyzed. In this case, when the synonym is converted and analyzed, it may be the keyword with the largest number, but when analyzed separately, both keywords may not be the keyword with the largest number. Accordingly, the processor 220 may make the analysis result more accurate by unifying the synonym into one keyword.

In this case, the processor 220 may convert a synonym into one keyword according to a preset keyword. For example, the processor 220 may be preset to unify synonyms of family members and family members into a family, and accordingly, the processor 220 may convert all family keywords into family keywords.

Also, the processor 220 may convert synonyms by unifying keywords with a larger number of synonyms.

5 is a diagram illustrating a document word matrix according to an embodiment of the present invention.

Referring to FIG. 5 , (a) may indicate a document word matrix of a first type, and (b) may indicate a document word matrix of a second type.

The processor 220 may generate a document word matrix using the extracted nouns. The document word matrix may be a matrix expressing the number of respective nouns disclosed in a plurality of documents. That is, the processor 220 may generate a document word matrix based on the number of each of the extracted nouns disclosed in each of the collected documents.

The row of the document word matrix of the first form may be the number of extracted nouns each appearing in a specific document, and the column may be the number of extracted nouns starting in each of the documents. That is, the number of rows of the document word matrix of the first form may be the number of extracted nouns, and the number of columns may be the number of collected documents.

The row of the document word matrix of the second form may be the number of extracted nouns in which a specific noun is disclosed in each document, and the column may be the number of each of the extracted nouns appearing in the specific document. That is, the number of rows of the document word matrix of the second form may be the number of collected documents, and the number of columns may be the number of extracted nouns.

6 is a diagram illustrating a keyword map according to an embodiment of the present invention.

Referring to FIG. 6 , the processor 220 may calculate a distance between keywords corresponding to each element of the document word matrix, and generate a keyword map based on the calculated distance.

The processor 220 may calculate a distance between keywords, classify keywords having a distance from a specific keyword within a predetermined distance for each layer, and generate a keyword map by connecting the keywords for each layer. That is, as a result of calculating the distance between keywords, keywords having a distance difference of N with respect to a specific keyword may be classified into the same layer. For example, keywords having a distance difference of N by N from keyword 1 may be keyword 2 and keyword 4. The processor 220 may classify keyword 2 and keyword 4 into the same layer.

Also, the processor 220 may classify keywords having a distance difference of M with respect to a specific keyword into the same layer. For example, a keyword having a distance difference between keyword 1 and M by M may be keyword 3.

Also, the processor 220 may classify keywords having a distance difference of K with respect to a specific keyword into the same layer. For example, the keyword having a distance difference of K by K from keyword 1 may be keyword 5.

The processor 220 may generate a keyword tree by connecting each layer according to a distance from a specific keyword.

In addition, the processor 220 calculates the distance between the keywords to connect the keywords having a distance from the specific keyword within a predetermined distance to the specific keyword, and connects keywords having a distance from each of the keywords within a predetermined distance to the keyword within a predetermined distance. can connect with That is, the processor 220 may generate a keyword map by connecting keywords having a close distance to the specific keyword with the specific keyword, and connecting keywords having a close distance to each of the keywords connected to the specific keyword. For example, as a result of calculating the distance between the keywords, the processor 220 may connect the keywords 2 and 4, which are keywords having a distance difference within N, with respect to the keyword 1, which is a specific keyword. Also, the processor 220 may connect keyword 2 and keyword 4 and keyword 3, which is a keyword having a distance difference within M, with keyword 2 and keyword 4 . Also, the processor 220 may connect keyword 3 and keyword 5, which is a keyword having a distance difference within K.

Here, the keywords having a distance difference within M from the keywords 2 and 4 are defined as one keyword 3, but the present invention is not limited thereto, and may be connected to a plurality of keywords. In addition, keywords having a distance difference within M of keyword 2 and keyword 4 may be different, and each of the keywords having a distance difference within M may be connected to at least one of keyword 2 or keyword 4 .

The processor 220 may display the keyword map on the output device. Accordingly, the user may easily identify related keywords in the plurality of documents.

7 is a diagram illustrating sorting of keywords according to an embodiment of the present invention.

Referring to FIG. 7 , (a) is the number of nouns from which rows are extracted each is disclosed in a specific document, and the number of specific nouns among nouns from which columns are extracted is the number of nouns that are disclosed in each document in the first form In the case of a matrix, (b) is the number of nouns from which a specific noun is disclosed in each document among the nouns from which the row is extracted, and the column is the number of each of the nouns from which the row is extracted is disclosed in a specific document. It can represent the case of a matrix.

The processor 220 may sequentially sort the keywords from the largest number of keywords to the smallest number of keywords in the document word matrix. The processor 220 may sort by changing only the order of rows or columns while maintaining the form of the document word matrix. For example, in the case of the document word matrix of the first type, the number of each noun appearing in the documents may be described in a row. The number of each keyword disclosed in the documents may be 8 for keyword 1, 9 for keyword 2, 15 for keyword 3, 10 for keyword 4, and 2 for keyword 5. Accordingly, when the keywords are arranged in the order of the large number of keywords, the order may be keyword 3 - keyword 4 - keyword 2 - keyword 1 - keyword 5. Accordingly, the processor 220 may change the order of the rows in which the number of keywords disclosed in the documents is described. Accordingly, the processor 220 may sort the keywords in the order from the keyword with the largest number to the keyword with the smallest number.

In addition, in the case of the document word matrix of the second type, the number of each noun that appears in the documents may be described in a column. The number of each keyword disclosed in the documents may be 8 for keyword 1, 9 for keyword 2, 15 for keyword 3, 10 for keyword 4, and 2 for keyword 5. Accordingly, when the keywords are arranged in the order of the large number of keywords, the order may be keyword 3 - keyword 4 - keyword 2 - keyword 1 - keyword 5. Accordingly, the processor 220 may change the order of columns in which the number of keywords disclosed in documents is described. Accordingly, the processor 220 may sort the keywords in the order from the keyword with the largest number to the keyword with the smallest number.

Here, the processor 220 may calculate weights according to the number of keywords included in the documents, respectively, according to the weights of the documents, and according to the calculated weights, the keywords are arranged in the order of the keywords from the largest number to the smallest. You can also sort.

The processor 220 may display the sorted keywords on the output device. Accordingly, the user may easily identify main keywords in the plurality of documents.

8 is a diagram illustrating sorting of sets according to an embodiment of the present invention.

Referring to FIG. 8 , the processor 220 may configure a set of keywords corresponding to each element of the document word matrix for each similar keyword, calculate a distance between the configured sets, and arrange the configured sets according to the distance. Here, the size of each circle may be different according to the number of classified keywords, and the distance between the circles may indicate the distance between the keywords.

The processor 220 may display the sorted sets on the output device. Accordingly, the user may easily identify similar keywords in a plurality of documents.

9 is a diagram illustrating a method for analyzing trends in airport and aviation technology according to an embodiment of the present invention.

Referring to FIG. 9 , the processor 220 may collect documents related to airport and aviation technology from the web 100 ( S100 ). The processor 220 may collect documents from news, academic, R&D projects, patents, and policy sites. The documents that the processor 220 collects from the web 100 may be documents related to airport and aviation technology. In addition, the processor 220 may filter documents about airports and aviation technologies among the documents collected from the web 100 , and store the filtered documents about airports and aviation technologies.

The processor 220 may extract text from documents collected from the web 100 and identify morphemes from the extracted text ( S200 ). The processor 220 may extract text from documents collected through text mining. The processor 220 may identify the morpheme from the extracted text by comparing the extracted text with the morphemes stored in the morpheme dictionary, respectively. The processor 220 may identify the same text as the morpheme stored in the morpheme dictionary from each extracted text as a morpheme.

The processor 220 may extract a noun by deleting a stopword from among the identified morphemes and converting a synonym from among the identified morphemes into one morpheme ( S300 ). Here, the stopwords may include words, articles, prepositions, articles, conjunctions, etc. that are not used as terms. That is, since an article, a preposition, a proposition, a conjunction, and the like are deleted from the morpheme identified by the processor 220 , only the morpheme in the form of a noun may remain. Accordingly, the processor 220 may extract a noun from among the identified morphemes. Also, the processor 220 may unify the synonyms by converting the synonyms into one morpheme.

The processor 220 may generate a document word matrix using the extracted nouns ( S400 ). The document word matrix may be a matrix expressing the number of respective nouns disclosed in a plurality of documents. That is, the processor 220 may generate a document word matrix based on the number of each of the extracted nouns disclosed in each of the collected documents.

The document word matrix may be generated in at least one of the first form and the second form. In the document word matrix of the first form, each of nouns from which rows are extracted may be the number of occurrences in a specific document, and among the nouns from which columns are extracted, the number of specific nouns may be disclosed in each of the documents. Also, in the document word matrix of the first form, the number of rows may be the number of extracted nouns, and the number of columns may be the number of documents collected.

In addition, the document word matrix of the second form may be the number of nouns from which rows are extracted from among the nouns that are disclosed in each document, and the number of nouns from which columns are extracted from each of the nouns that are disclosed in the specific document. Also, in the document word matrix of the second form, the number of rows may be the number of documents collected, and the number of columns may be the number of extracted nouns.

The processor 220 may analyze the degree of similarity and relevance between keywords corresponding to each element of the document word matrix and the degree of similarity and relevance between documents using the document word matrix, and may provide the analyzed result to the user ( S500).

The processor 220 may calculate a distance between keywords corresponding to each element of the document word matrix, and generate a keyword map based on the calculated distance.

The processor 220 may calculate a distance between keywords, classify keywords having a distance from a specific keyword within a predetermined distance for each layer, and generate a keyword map by connecting the keywords for each layer.

In addition, the processor 220 calculates the distance between the keywords to connect the keywords having a distance from the specific keyword within a predetermined distance to the specific keyword, and connects keywords having a distance from each of the keywords within a predetermined distance to the keyword within a predetermined distance. can connect with That is, the processor 220 may generate a keyword map by connecting keywords having a close distance to the specific keyword with the specific keyword, and connecting keywords having a close distance to each of the keywords connected to the specific keyword.

The processor 220 may display the generated keyword map on an output device and provide the analyzed result to the user.

Also, the processor 220 may sequentially sort the keywords from the largest number to the smallest keyword in the document word matrix. The processor 220 may sort by changing only the order of rows or columns while maintaining the form of the document word matrix.

Also, the processor 220 may separately sort the keywords in an order of increasing the number of keywords. For example, the processor 220 may sort keywords in the form of (keyword 1, keyword 2, keyword 3, ??).

The processor 220 may display the sorted keywords on the output device to provide the analyzed result to the user.

Also, the processor 220 may configure a set of keywords corresponding to each element of the document word matrix for each similar keyword, calculate a distance between the configured sets, and arrange the configured sets according to the distance.

The processor 220 may provide the analyzed result to the user by displaying the sorted sets on the output device.

As described above, according to an embodiment of the present invention, data on airport and aviation technology is filtered from data collected from web content, keywords are extracted by analyzing morphemes from the filtered data, and correlation between the extracted keywords is analyzed. Thus, it is possible to realize the trend analysis device and method of airport and aviation technology to analyze trends.

Those skilled in the art to which the present invention pertains should understand that the present invention may be embodied in other specific forms without changing the technical spirit or essential characteristics thereof, so the embodiments described above are illustrative in all respects and not restrictive. only do The scope of the present invention is indicated by the following claims rather than the detailed description, and all changes or modifications derived from the meaning and scope of the claims and their equivalent concepts should be construed as being included in the scope of the present invention. .

100: web
200: Trend analysis device of airport and aviation technology
210: memory
220: processor

Claims (26)

collecting documents related to airport and aviation technology from the web;
extracting text from the collected documents and identifying morphemes from the extracted text;
extracting a noun from among the identified morphemes;
generating a document word matrix using the extracted nouns; and
airports and airlines, comprising the steps of using the document word matrix to analyze the degree of similarity and relevance between keywords corresponding to each element of the document word matrix, and the degree of similarity and relevance between each document, and providing an analysis result How to analyze trends in technology. According to claim 1,
Collecting the documents includes:
Airport to filter documents related to airport and aviation technology using the similarity between the address of the site, the name of the site, the name of the document, the content of the document, the author of the document, the host of the site, and the words related to the airport and aviation technology. and trend analysis methods in aviation technology. According to claim 1,
Collecting the documents includes:
A trend analysis method of airport and aviation technology that collects representative keywords of each document by analyzing HTML tags of news, academics, R&D projects, patents and policy sites published on the web. 4. The method of claim 3,
A morpheme dictionary in which morphemes related to airport and aviation technology are stored is stored in the memory,
Collecting documents from the web includes:
A trend analysis method of airport and aviation technology for updating the morpheme dictionary based on the collected representative keywords. 5. The method of claim 4,
The step of identifying the morpheme in the extracted text includes:
A trend analysis method of airport and aviation technology for identifying morphemes in the extracted text by comparing the extracted text with the morphemes stored in the morpheme dictionary. According to claim 1,
The step of extracting the noun is
A trend analysis method of airport and aviation technology that deletes stopwords among identified morphemes and extracts nouns by converting synonyms among identified morphemes into one morpheme. According to claim 1,
The step of generating the document word matrix comprises:
A trend analysis method of airport and aviation technology for generating the document word matrix by converting the number of extracted nouns respectively disclosed in each document into a matrix. According to claim 1,
The step of generating the document word matrix comprises:
The document word matrix is generated in at least one of a first form and a second form,
In the first form, each of the nouns from which the row is extracted is the number included in a specific document, and the specific noun among the nouns from which the column is extracted is the number included in each of the documents,
The second form is the number of nouns from which columns are extracted each included in a specific document, and the number of specific nouns among nouns from which rows are extracted are included in each of the documents. A method for analyzing trends in airport and aviation technology. According to claim 1,
The step of generating the document word matrix comprises:
The document word matrix is generated in at least one of a first form and a second form,
In the first form, the number of rows is the number of extracted nouns, the number of columns is the number of documents collected,
In the second form, the number of rows is the number of documents collected, and the number of columns is the number of extracted nouns. A method for analyzing trends in airport and aviation technology. According to claim 1,
The step of providing the analysis result is,
An airport and aviation technology trend analysis method for calculating a distance between keywords corresponding to each element of the document word matrix, generating a keyword map based on the calculated distance, and providing the generated keyword map. 11. The method of claim 10,
The step of providing the analysis result is,
A trend analysis method of airport and aviation technology that classifies keywords that are within a certain distance from a specific keyword by layer, and creates a keyword map by connecting keywords for each layer. According to claim 1,
The step of providing the analysis result is,
A trend analysis method in airport and aviation technology for sequentially arranging keywords corresponding to each element of the document word matrix in the order from the keyword with the largest number to the keyword with the smallest number, and providing the sorted keywords. According to claim 1,
The step of providing the analysis result is,
A method for analyzing trends in airport and aviation technology that configures a set of keywords corresponding to each element of the document word matrix for each similar keyword, calculates the distance between the configured sets, sorts sets according to the distance, and provides the sorted sets . a memory for storing documents collected from the web related to airport and aviation technology; and
Extracting text from the collected documents, identifying morphemes from the extracted text, extracting nouns from among the identified morphemes, generating a document word matrix using the extracted nouns, and using the document word matrix A trend analysis apparatus for airport and aviation technology, comprising: a processor for analyzing the degree of similarity and relevance between keywords corresponding to each element of the document word matrix, and the degree of similarity and relevance between each document; and providing an analysis result. 15. The method of claim 14,
The processor uses the address of the site, the name of the site, the name of the document, the content of the document, the author of the document, and the similarity between each host of the site and the words related to airport and aviation technology to generate documents related to airport and aviation technology. Filtering airport and aviation technology trend analysis device. 15. The method of claim 14,
The memory is an airport and aviation technology trend analysis device for storing representative keywords of each document collected by analyzing HTML tags of news, academic, R&D projects, patents and policy sites disclosed on the web. 17. The method of claim 16,
A morpheme dictionary in which morphemes related to airport and aviation technology are stored is stored in the memory;
An airport and aviation technology trend analysis device in which the morpheme dictionary is updated based on the collected representative keywords. 18. The method of claim 17,
The processor compares the extracted text with the morphemes stored in the morpheme dictionary to identify the morpheme from the extracted text. 15. The method of claim 14,
The processor deletes stopwords among the identified morphemes, and converts synonyms among the identified morphemes into one morpheme to extract nouns. 15. The method of claim 14,
The processor converts the number of each extracted nouns disclosed in each document into a matrix to generate the document word matrix. 15. The method of claim 14,
The document word matrix is generated in at least one of a first form and a second form,
In the first form, each of the nouns from which the row is extracted is the number included in a specific document, and the specific noun among the nouns from which the column is extracted is the number included in each of the documents,
In the second form, each of the nouns from which the column is extracted is the number included in a specific document, and the number of specific nouns among the nouns from which the row is extracted is included in each of the documents. 15. The method of claim 14,
The document word matrix is generated in at least one of a first form and a second form,
In the first form, the number of rows is the number of extracted nouns, the number of columns is the number of documents collected,
In the second form, the number of rows is the number of collected documents, and the number of columns is the number of extracted nouns. An apparatus for analyzing trends in airport and aviation technology. 15. The method of claim 14,
The processor calculates a distance between keywords corresponding to each element of the document word matrix, generates a keyword map based on the calculated distance, and provides the generated keyword map. 24. The method of claim 23,
The processor classifies keywords that are within a certain distance from a specific keyword by layer, and connects the keywords for each layer to generate a keyword map. 15. The method of claim 14,
The processor sequentially sorts the keywords corresponding to each element of the document word matrix in order from the keyword with the largest number to the keyword with the smallest number, and provides the sorted keywords. 15. The method of claim 14,
The processor configures a set of keywords corresponding to each element of the document word matrix for similar keywords, calculates the distance between the configured sets, sorts sets according to the distance, and provides sorted sets of airport and aviation technology. trend analysis device.

Images