빅데이터 기술에 대한 관심이 급증함에 따라, 소셜 미디어를 통해 유통되는 방대한 양의 비정형 데이터를 분석하고자 하는 시도가 활발히 이루어지고 있다. 이에 따라서 텍스트 형태의 비정형 데이터 분석을 통해 의미 있는 정보를 찾고자 하는 시도가 비즈니스 영역뿐 아니라, 정치, 경제, 문화 등 다양한 영역에서 이루어지고 있다. 특히 최근에는 여러 현안 및 이슈들을 발굴하여 이를 의사결정에 활용하고자 하는 시도가 활발히 이루어지고 있다. 이처럼 빅데이터 분석을 통해 국가현안이나 이슈를 발굴하고자 하는 시도가 꾸준히 이루어져왔음에도 불구하고, 국가현안 및 이슈로부터 이와 관련된 R&D 문서를 효율적으로 제공하는 방안은 마련되지 않고있다. 이는 사용자들이 인식하는 현안 키워드와 실제 사용되는 R&D 키워드 사이의 이질성이 존재하기 때문이다. 따라서 현안 및 R&D키워드간의 이질성을 극복하기 위한 중간 장치가 필요하며, 이 중간 장치를 통해 각 현안 키워드와 R&D 키워드간에 적절한 대응이 이루어져야 한다. 이를 위해 본 연구에서는 (1) 현안 키워드 추출을 위한 하이브리드 방법론, (2) 현안 대응 R&D 정보 패키징 방법론, 그리고 (3) R&D 관점에서의 연관 현안 네트워크 구축 방법론의 총 세 가지 방법론을 제안한다. 제안하는 방법론은 텍스트 마이닝, 소셜네트워크 분석, 그리고 연관 규칙 마이닝 등의 데이터 분석 기법들을 활용하여 수행하였으며, 그 결과, (1)에 의한 키워드 보강률은 42.8%로 나타났으며, (2)의 경우, 현안 키워드와 R&D 키워드간 다수의 연관 규칙이 나타났다. (3)의 경우는 현재 진행 중에 있으며, 향후 가시적 성과를 낼 수 있을 것으로 예상된다.
빅데이터 기술에 대한 관심이 급증함에 따라, 소셜 미디어를 통해 유통되는 방대한 양의 비정형 데이터를 분석하고자 하는 시도가 활발히 이루어지고 있다. 이에 따라서 텍스트 형태의 비정형 데이터 분석을 통해 의미 있는 정보를 찾고자 하는 시도가 비즈니스 영역뿐 아니라, 정치, 경제, 문화 등 다양한 영역에서 이루어지고 있다. 특히 최근에는 여러 현안 및 이슈들을 발굴하여 이를 의사결정에 활용하고자 하는 시도가 활발히 이루어지고 있다. 이처럼 빅데이터 분석을 통해 국가현안이나 이슈를 발굴하고자 하는 시도가 꾸준히 이루어져왔음에도 불구하고, 국가현안 및 이슈로부터 이와 관련된 R&D 문서를 효율적으로 제공하는 방안은 마련되지 않고있다. 이는 사용자들이 인식하는 현안 키워드와 실제 사용되는 R&D 키워드 사이의 이질성이 존재하기 때문이다. 따라서 현안 및 R&D키워드간의 이질성을 극복하기 위한 중간 장치가 필요하며, 이 중간 장치를 통해 각 현안 키워드와 R&D 키워드간에 적절한 대응이 이루어져야 한다. 이를 위해 본 연구에서는 (1) 현안 키워드 추출을 위한 하이브리드 방법론, (2) 현안 대응 R&D 정보 패키징 방법론, 그리고 (3) R&D 관점에서의 연관 현안 네트워크 구축 방법론의 총 세 가지 방법론을 제안한다. 제안하는 방법론은 텍스트 마이닝, 소셜네트워크 분석, 그리고 연관 규칙 마이닝 등의 데이터 분석 기법들을 활용하여 수행하였으며, 그 결과, (1)에 의한 키워드 보강률은 42.8%로 나타났으며, (2)의 경우, 현안 키워드와 R&D 키워드간 다수의 연관 규칙이 나타났다. (3)의 경우는 현재 진행 중에 있으며, 향후 가시적 성과를 낼 수 있을 것으로 예상된다.
Considerable research efforts are being directed towards analyzing unstructured data such as text files and log files using commercial and noncommercial analytical tools. In particular, researchers are trying to extract meaningful knowledge through text mining in not only business but also many othe...
Considerable research efforts are being directed towards analyzing unstructured data such as text files and log files using commercial and noncommercial analytical tools. In particular, researchers are trying to extract meaningful knowledge through text mining in not only business but also many other areas such as politics, economics, and cultural studies. For instance, several studies have examined national pending issues by analyzing large volumes of text on various social issues. However, it is difficult to provide successful information services that can identify R&D documents on specific national pending issues. While users may specify certain keywords relating to national pending issues, they usually fail to retrieve appropriate R&D information primarily due to discrepancies between these terms and the corresponding terms actually used in the R&D documents. Thus, we need an intermediate logic to overcome these discrepancies, also to identify and package appropriate R&D information on specific national pending issues. To address this requirement, three methodologies are proposed in this study-a hybrid methodology for extracting and integrating keywords pertaining to national pending issues, a methodology for packaging R&D information that corresponds to national pending issues, and a methodology for constructing an associative issue network based on relevant R&D information. Data analysis techniques such as text mining, social network analysis, and association rules mining are utilized for establishing these methodologies. As the experiment result, the keyword enhancement rate by the proposed integration methodology reveals to be about 42.8%. For the second objective, three key analyses were conducted and a number of association rules between national pending issue keywords and R&D keywords were derived. The experiment regarding to the third objective, which is issue clustering based on R&D keywords is still in progress and expected to give tangible results in the future.
Considerable research efforts are being directed towards analyzing unstructured data such as text files and log files using commercial and noncommercial analytical tools. In particular, researchers are trying to extract meaningful knowledge through text mining in not only business but also many other areas such as politics, economics, and cultural studies. For instance, several studies have examined national pending issues by analyzing large volumes of text on various social issues. However, it is difficult to provide successful information services that can identify R&D documents on specific national pending issues. While users may specify certain keywords relating to national pending issues, they usually fail to retrieve appropriate R&D information primarily due to discrepancies between these terms and the corresponding terms actually used in the R&D documents. Thus, we need an intermediate logic to overcome these discrepancies, also to identify and package appropriate R&D information on specific national pending issues. To address this requirement, three methodologies are proposed in this study-a hybrid methodology for extracting and integrating keywords pertaining to national pending issues, a methodology for packaging R&D information that corresponds to national pending issues, and a methodology for constructing an associative issue network based on relevant R&D information. Data analysis techniques such as text mining, social network analysis, and association rules mining are utilized for establishing these methodologies. As the experiment result, the keyword enhancement rate by the proposed integration methodology reveals to be about 42.8%. For the second objective, three key analyses were conducted and a number of association rules between national pending issue keywords and R&D keywords were derived. The experiment regarding to the third objective, which is issue clustering based on R&D keywords is still in progress and expected to give tangible results in the future.
* AI 자동 식별 결과로 적합하지 않은 문장이 있을 수 있으니, 이용에 유의하시기 바랍니다.
문제 정의
The third objective was to identify a set of related issues from an R&D perspective.
Therefore, the second goal of this study is to establish a methodology for packaging R&D information that corresponds to national issues.
Unfortunately, data-driven methods cannot distinguish between keywords pertaining to national pending issues (national pending issue keywords) and simple gossip keywords. Thus, the first goal of this study is to establish a hybrid approach for extracting and integrating national issue keywords in order to overcome the limitations of the top-down and data-driven approaches.
제안 방법
Third, it is necessary to automatize the human intervention at each step of the analysis in order to improve the applicability of the proposed methodology. Finally, more precise validation and verification based on mathematical theory are required in order to apply the proposed methodology.
Since the purpose of this stage is to achieve issue clustering from an R&D perspective, it essential to first construct the R&D lexicon based on R&D documents such as patent information or research reports.
The proposed approach is designed such that it can be applied to one of the core modules of the methodology for packaging R&D information services regarding issue.
The second objective was to address the discrepancies between the general issue keywords recognized by users and R&D keywords as well as package appropriate R&D information pertaining to specific issues.
However, these methods cannot distinguish between keywords pertaining to issue keywords and general gossip keywords. Therefore, in this study, we propose a hybrid approach for selecting and integrating issue keywords in order to overcome the limitations of the top-down and data-driven approaches. The proposed approach is designed such that it can be applied to one of the core modules of the methodology for packaging R&D information services regarding issue.
Third, it is necessary to automatize the human intervention at each step of the analysis in order to improve the applicability of the proposed methodology. Finally, more precise validation and verification based on mathematical theory are required in order to apply the proposed methodology.
To address the second objective of this study, three key analyses were conducted and a number of association rules between national pending issue keywords and R&D keywords were derived.
대상 데이터
The documents that contain both issue keywords and R&D keywords were collected for this study. Among the collected documents, documents provided by the NTIS (National Technical Information Service) were selected for the analysis. From 28,060 research reports provided by NTIS, only the main part of the research reports that were published from 2010–2012 are considered.
An analysis of the degree of enhancement of keywords is performed for each of the 140 themes. In the experiment, 1,148 issue keywords and 1,212 general issue keywords were used. Among the 1,212 general issue keywords, 858 keywords were involved in the integrated list and others were ignored.
In this study, 140 issue themes and 1,148 seed keywords were extracted from standard documents published by the Korean government. Further, additional topics were obtained from 11,000 documents through topic analysis using SAS Enterprise Miner.
Further, additional topics were obtained from 11,000 documents through topic analysis using SAS Enterprise Miner. The final list of integrated issue keywords consisted of 1,148 keywords from standard documents and 1,212 keywords from other documents such as policy reports, news articles, discussions, and columns.
In this process, a lexicon on national pending issue and a temporary R&D lexicon were used to eliminate the terms that are irrelevant to national pending issue or R&D. The issue lexicon is constructed based on the topic analysis of 11,160 cases that are selected from a variety of documents such as policy documents, news articles, discussions, and columns. And then, the issue lexicon and the R&D lexicon are combined.
참고문헌 (28)
R. Albright, "Taming Text with the SVD," SAS Institute Inc., 2004. www.sas.com/apps/whitepapers/whitepaper.jsp?codeSDM5
J. Han, and M. Kamber, "Data Mining: Concepts and Techniques," 3rd ed., Morgan Kaufmann Publishers: Massachusetts, 2011. http://web.engr.illinois.edu/-hanj/bk3/
R. J. Mooney, and R. Bunescu, "Mining Knowledge from Text Using Information Extraction," ACM SIGKDD Explorations, Vol. 7, No. 1, pp. 3-10, 2005. http://www.cs.utexas.edu/-ai-lab/pubs/text-kddexplore-05.pdf
I. H. Witten, "Text Mining," Practical Handbook of Internet Computing, edited by M. P. Singh, Chapman & Hall/ CRC Press, 2005. http://www.cs.waikato.ac.nz/-ihw/papers/04-IHW-Textmining.pdf
G. Salton, A. Wong, and C. S. Yang, "A Vector Space Model for Automatic Indexing," Communications of the ACM, Vol. 18, No. 11, pp. 613-620, 1975. http://dx.doi.org/10.1145/361219.361220
A. Stanvrianou, P. Andritsos, and N. Nicoloyannis, "Overview and Semantic Issues of Text Mining," ACM SIGMOD Record, Vol. 36, No. 3, pp. 23-34, 2007. http://dx.doi.org/10.1145/1324185.1324190
E. Yu, J. Kim, C. Lee, and N. Kim, "Using Ontologies for Semantic Text Mining," The Journal of Information Systems, Vol. 21, No. 3, pp. 137-161, 2012. http://dx.doi.org/10.5859/KAIS.2012.21.3.137
D. Jeong, M. Hwang, M. Cho, H. Jung, S. Yoon, K. Kim, and P. Kim, "Ontology and Text Mining-based Advanced Historical People Finding Service," Journal of Internet Computing and Services, Vol. 13, No. 5, pp. 33-43, 2012. http://dx.doi.org/10.7472/jksii.2012.13.5.33
R. Agrawal, and R. Srikant, "Fast Algorithms for Mining Association Rules," International Conference on Very Large Data Bases, Santiago, Chile, pp. 487-499, 1994. http://dl.acm.org/citation.cfm?id645920.672836&collDL&dlACM&CFID656631652&CFTOKEN27677818
I. Cho, and N. Kim, "Recommending Core and Connecting Keywords of Research Area Using Social Network and Data Mining Techniques," Journal of Intelligence and Information Systems, Vol. 17, No. 1, pp. 127-138, 2011. http://www.dbpia.co.kr/Journal/ArticleDetail/1477011
Y. Sohn, I. Kim, and N. Kim, "Automated Conceptual Data Modeling Using Association Rule Mining," The Journal of information systems, Vol. 18, No. 4, pp. 59-86, 2009. http://dx.doi.org/10.5859/KAIS.2009.18.4.059
N. Kim, "Effect of Market Basket Size on the Accuracy of Association Rule Measures," The journal of MIS research, Vol. 18, No. 2, pp. 95-114, 2008. http://scholar.ndsl.kr/schDetail.do
H. Ahn, I. Han, and N. Kim, "The Product Recommender System Combining Association Rules and Classification Models: The Case of G Internet Shopping Mall," Information Systems Review, Vol. 8, No. 1, pp. 181-201, 2006. http://dx.doi.org/10.13088/jiis.2013.19.2.039
S. Yoon, "Churn Prediction Model for Department Store Customers Using Data Mining Technique," Asia Marketing Journal, Vol. 6, No. 4, pp. 45-72, 2005. http://academic.naver.com/view.nhn?doc_id11465855
Y. Lee, and K. Kim, "Product Recommender Systems using Multi-Model Ensemble Techniques," Journal of Intelligence and Information Systems, Vol. 19, No. 2, pp. 39-54, 2013. http://www.dbpia.co.kr/Article/3219909
W. F. Wang, Y. L. Chung, M. H. Hus, and A. C. Keh, "A Personalized Recommender System for the Cosmetic Business," Expert Systems with Applications, Vol. 26, No. 3, pp. 427-434, 2007. http://dx.doi.org/10.1016/j.eswa.2003.10.001
Y. Kim, "Social Network Analysis," Bakyoungsa: Seoul, 2003. http://book.naver.com/bookdb/book_detail.nhn?bid128306
S. Kauffman, "The Origins of Order," Oxford University Press: New York, 1993. https://global.oup.com/academic/product/the-origins-of-order-9780195079517
K. Kwahk, "Social Network Analysis," Chungram: Seoul, 2013. http://book.naver.com/bookdb/book_detail.nhn?bid7462254
S. Park, and K. P. Kim, "A Closeness Analysis Algorithm for Workflow-supported Social Networks," Journal of Internet Computing and Services, Vol. 14, No. 5, pp. 77-85, 2013. http://www.dbpia.co.kr/Article/3282313
K. Lee, H. Namgoong, E. Kim, K. Lee, and H. Kim, "Analysis of Multi-Dimensional Interaction among SNS Users," Journal of Internet Computing and Services, Vol. 12, No. 2, pp. 113-122, 2011. http://www.dbpia.co.kr/Article/1464198
A. Jin, J. Lee, and J. Lee, "Measuring Method of String Similarity for POI Data Retrieval," Journal of KIISE: Computing Practices and Letters, Vol. 19, No. 4, pp. 177-185, 2013. http://www.dbpia.co.kr/Article/3140094
B. You, and K. Choi, "A Study on the Construction of the National R&D Knowledge Information: Mainly Focused on the Research Planning and Management," Journal of the Korean Society for Library and Information Science, Vol. 38, No. 1, pp. 281-301, 2004. http://www.dbpia.co.kr/Article/348410
S. Shin, Y. Yoon, M. Yang, J. Kim, and K. Shon, "A Data Cleansing Strategy for Improving Data Quality of National R&D Information-Case Study of NTIS," The Korean Society Of Computer And Information, Vol. 16, No. 6, 2011. http://dx.doi.org/10.9708/jksci.2011.16.6.119
Y. Hyun, H. Han, H. Choi, J. Park, K. Lee, K. Kwahk, and N. Kim, "Methodology Using Text Analysis for Packaging R&D Information Services on Pending National Issues," Journal of Information Technology Applications & Management, Vol. 20, No. 3, pp. 231-257, 2013. http://www.dbpia.co.kr/Article/3257838
L. Kwon, and J. Kim, "A Study on the Establishment of Reference Linking System for National R&D Information," Journal of Korea Contents Association, Vol. 8, No. 1, pp. 195-202, 2008. http://www.dbpia.co.kr/Article/761180
M. Yang, Y. Yoon, S. Shin, J. Kim, and K. Shon, "A Development of Expert Search Agent System using National R&D Human Information Database for NTIS," Journal of Internet Computing and Services, Vol. 11, No. 2, pp. 285-286, 2010. http://www.dbpia.co.kr/Article/1390628
J. Scott, Social Network Analysis: A Handbook, SAGE: California, 2000. http://www.amazon.com/Social-Network-Analysis-John-Scott/dp/1446209040
※ AI-Helper는 부적절한 답변을 할 수 있습니다.