참고문헌의 유무가 LDA를 사용한 토픽 모델링 결과에 미치는 영향 = The Effect of Reference List in the Article on Topic Modeling using LDA

국문 초록 (Abstract)

본 연구는 논문을 사용한 토픽 모델링에서 연구자가 고려해야 할 요소 중 하나인 참고문헌 제거 여부에 대한 필요성을 탐색하고자 하였다. 참고문헌 내용만을 텍스트에서 정확하게 제거하는 명령어나 기술이 제공되고 있지 않아, 참고문헌을 제외한 텍스트로 토픽 모델링을 수행하고자 하였을 때에는 연구자가 모든 텍스트에서 일일이 참고문헌을 제거하는 과정을 거쳐야만 한다. 이렇게 많은 노동력과 시간이 소요되는 참고문헌 제거 과정이 연구자에게 부담으로 작용할 수 있으므로, 참고문헌의 존재가 토픽 모델링 결과에 어떠한 영향을 미치는지를 알아보고 토픽 모델링에 더욱 적합한 텍스트 형태를 제시하고자 하였다. 스포츠산업경영 분야의 국제 저널 SMQ, ESMQ와 JSM에서 최근 10년간 출간된 논문을 대상으로 토픽 모델링을 수행하고, 참고문헌이 포함된 텍스트와 참고문헌이 제거된 텍스트의 토픽 모델링 결과를 비교하였다. 스포츠산업경영 및 마케팅에 대한 전반적인 지식을 가지고 세 가지 저널의 특성에 대해 충분히 이해하고 있는 전문가 5인의 평가로 결과 비교가 이루어졌다. 비교 결과, 참고문헌을 포함한 전문을 사용하였을 때 더욱 해석이 용이한 토픽 모델링 결과가 도출되는 것으로 나타났다. 텍스트에 참고문헌이 포함되었을 때에 토픽을 산출하는 데 할당되는 논문 간의 연관성이 높았으며, 연관성이 높은 논문들로부터 산출된 토픽의 해석가능성 또한 높아지는 것으로 판단할 수 있다. 이처럼 참고문헌의 존재가 토픽 모델링 결과에 부정적인 영향을 미치지 않는다는 본 연구결과에 따르면, 논문을 사용한 토픽 모델링에서 참고문헌을 제거하는 추가적인 절차를 거치지 않아도 됨을 뜻한다. 이에 따라, 논문을 사용한 토픽 모델링을 수행하고자 하는 추후 연구에서는 참고문헌의 존재가 미칠 수 있는 토픽 모델링 결과에 대한 부정적인 영향력에 대한 고민을 덜 수 있을 것으로 판단된다.

다국어 초록 (Multilingual Abstract)

This study tried to investigate the necessity of the reference removal procedure during the text pre-processing in topic modeling by comparing the topic modeling results using the text with reference and the text without reference. The use of full-text article for the topic modeling may cause extra work on removing the subsidiary part that is not directly related to the content of the article, because it is not able to selectively extract the certain part of the text that researcher wants to include in the analysis has not yet been developed. Therefore, manual work by the researcher is required to remove the subsidiary information and it takes a great amount of labor and time to perform the text pre-processing rather than the actual topic modeling analysis process. Thus, this study compared the topic modeling result of the text containing reference and of the text without reference. The research articles has been published from three major sports management and marketing journals from 2009 to 2018 were collected. JSM, ESMQ, and SMQ have been selected based on their reputation and status in the academic field. Five sport management/marketing experts were hired to evaluate the quality of the topic modeling result. The result indicated that the text containing reference produced a better topic modeling result than the text without reference. The relevance between the articles which were used to generate certain topic was higher when the text containing reference. It can be concluded that the quality of the topics is also high when the relevance of the articles is high. Based on this result, reference removal procedure is not necessary for the topic modeling using research articles. It can reduce a considerable amount of time and labor that the researcher needs to spend on pre-processing.

참고문헌 (Reference)

1 Paul, M. J., "You are what you tweet: Analyzing twitter for public health" 2011

2 Yu, Y., "World Cup 2014 in the Twitter World : A big data analysis of sentiments in US sports fans’ tweets" 48 : 392-400, 2015

3 Trappey, C., "Using the collective intelligence of sports fans to improve professional football league customer service" IEEE 313-318, 2014

4 Paul, M., "Topic modeling of research fields: An interdisciplinary perspective" 337-342, 2009

5 McCallum, A., "Topic and role discovery in social networks with experiments on enron and academic email" 30 : 249-272, 2007

6 Reed, J. G., "The psychology research handbook:a guide for graduate students and research assistants" 41-, 2006

7 Milojević, S., "The cognitive structure of library and information science : Analysis of article title words" 62 (62): 1933-1953, 2011

8 Westergaard, D., "Text mining of 15 million full-text scientific articles"

9 Blei, D. M., "Text Mining" Chapman and Hall/CRC 101-124, 2009

10 Ciomaga, B., "Sport management : A bibliometric study on central themes and trends" 13 (13): 557-578, 2013

11 Amado, A., "Research trends on Big Data in Marketing : A text mining and topic modeling based literature analysis" 24 (24): 1-7, 2018

12 Chandra, Y., "Mining social entrepreneurship strategies using topic modeling" 11 (11): e0151342-, 2016

13 Suominen, A., "Map of science with topic modeling : Comparison of unsupervised learning and human-assigned subject classification" 67 (67): 2464-2476, 2016

14 Chuang, J., "Large-scale examination of academic publications using statistical models" 2012

15 Quinn, K. M., "How to analyze political attention with minimal assumptions and costs" 54 (54): 209-228, 2010

16 Bornmann, L., "Growth rates of modern science : A bibliometric analysis based on the number of publications and cited references" 66 (66): 2215-2222, 2015

17 Syed, S., "Full-text or abstract? Examining topic coherence scores using latent dirichlet allocation" IEEE 165-174, 2017

18 Slack, T., "From the locker room to the board room : Changing the domain of sport management" 10 (10): 97-105, 1996

19 Griffiths, T. L., "Finding scientific topics" 101 (101): 5228-5235, 2004

20 Hong, L., "Empirical study of topic modeling in twitter" acm 80-88, 2010

21 Sun, L., "Discovering themes and trends in transportation research using topic modeling" 77 : 49-66, 2017

22 Zhao, W. X., "Comparing twitter and traditional media using topic models" Springer-Verlag Berlin Heidelberg 338-349, 2011

23 Rip, A., "Co-word maps of biotechnology : An example of cognitive scientometrics" 6 (6): 381-400, 1984

24 Yau, C. K., "Clustering scientific documents with topic modeling" 100 (100): 767-786, 2014

25 Guo, L., "Big social data analytics in journalism and mass communication : Comparing dictionary-based text analysis and unsupervised topic modeling" 93 (93): 332-359, 2016

26 Kurata, K., "Analyzing library and information science full‐text articles using a topic modeling approach" 55 (55): 847-848, 2018

27 Newman, D., "Analyzing entities and topics in news articles using statistical topic models" Springer-Verlag Berlin Heidelberg 93-104, 2006

28 Guerreiro, J., "A text mining-based review of cause-related marketing literature" 139 (139): 111-128, 2016

29 Kane, D. A., "A systematic review of perennial staple crops literature using topic modeling and bibliometric analysis" 11 (11): e0155788-, 2016

30 Du, L., "A segmented topic model based on the two-parameter Poisson-Dirichlet process" 81 (81): 5-19, 2010

31 Gatti, C. J., "A historical analysis of the field of or/ms using topic models"

32 Park, Y. J., "A deep learning-based sports player evaluation model based on game statistics and news articles" 138 : 15-26, 2017

33 Shilbury, D., "A bibliometric study of citations to sport management and marketing journals" 25 (25): 423-444, 2011

34 문화체육관광부, "2017 스포츠산업 백서" 문화체육관광부 2017