본문 바로가기
KERD
논문KCI 등재

AI 기반 맞춤형 논문 추천 시스템 개발 및 적용 사례 연구: K 대학을 중심으로

Development and Application of an AI-Based Personalized Research-Paper Recommendation System: An Example from K University

0

조회

0

다운로드

1

인용

0

참고문헌

키워드

주제분류교육학

초록

대학원생이 최신 연구 동향을 파악하고 본인의 연구 주제를 확립하는 것은 연구역량 강화에 있어서 중요하다. 본 연구는 대학원생의 연구역량 강화를 위해 AI 기반 맞춤형 논문 추천 시스템을 개발하고 이를 대학 현장에 적용하고자 하였다. 본 맞춤형 논문 추천 시스템은 콘텐츠 기반 필터링 알고리즘과 학문용어사전을 활용하여 대학원생의 관심 연구 키워드와 유사한 최신 논문을 추천한다. 본 연구는 TF-IDF(Term Frequency-Inverse Document Frequency) 알고리즘을 활용하여 교내 데이터베이스에 축적된 학생 관련 데이터로부터 학생의 관심 연구 키워드를 추출하였다. 맞춤형 논문 추천 시스템은 2021년 7월 파일럿 테스트를 시작으로 이메일 시스템을 통해 월 1회 대학원생에게 10개의 추천 논문을 제공하고 있다. 논문 추천 시스템의 개선을 목적으로 실시된 서비스 만족도 설문조사 결과, 학생들은 전반적으로 맞춤형 추천 서비스 이용 및 결과에 만족하는 것으로 나타났다. 또한, 논문 추천 결과와 추천 시스템에서 추출된 학생의 관심 연구 키워드에 대한 Precision@K는 각각 50.2%, 52.2%로 준수한 모델 성능을 보이는 것으로 나타났다. 본 연구는 맞춤형 논문 추천 시스템 개발을 위해 교육 및 연구 데이터를 통합하여 분석하였다는 점에서 의의가 있다. 또한, 많은 고등교육기관이 대학원생을 위한 AI 기반 맞춤형 교육 서비스를 개발하고 적용하고자 할 때 본 연구가 기본적인 지침의 역할을 하기를 기대한다.

Abstract

Graduate students need to identify the latest research trends and establish their own research topics to enhance their research competency. This study aimed to develop an AI-based personalized research-paper recommendation system to promote graduate students’ research competency and implement it in higher education. The recommendation system leveraged content-based filtering algorithms and an academic term dictionary to recommend recent research papers that are similar to graduate students’ research keywords of interest. We extracted their core keywords from student-related data stored in the university’s database utilizing TF-IDF(Term Frequency-Inverse Document Frequency) algorithms. Starting with a pilot test in July 2021, we provided graduate students with ten recommended papers every month via the university’s email service. As a result of the satisfaction survey on service use conducted to improve the recommendation system, students were generally satisfied with the use and recommendation results of the system. In addition, we found that Precision@K for the recommendation results and extracted keywords showed 50.2% and 52.2%, respectively. This study is significant because we performed integrated analyses of education and research data to develop a personalized research-paper recommendation system. Moreover, it can be a basic guide for other higher education institutions to create and apply AI-based personalized educational services for graduate students.

참고문헌 (38)

  1. 인공지능시대 교육정책방향과 핵심과제, "교육부", 교육부, ~ (2020)
  2. 김두형, "협업 필터링을 활용한대학 교양과목 추천 시스템", 대한산업공학회 추계학술대회 논문집, pp.2551~2556 (2020)
  3. 나승일, "산업교육분야 전공 대학원생의 연구역량 형성 과정", 농업교육과 인적자원개발, 41(3), pp.87~121 (2009), 한국농·산업교육학회[DOI]
  4. 신택수, "고등교육기관 핵심 역량의 구조화 및 경쟁력 지표의 개발과 적용: S대학교 교육학과 대학원 과정을 중심으로", 아시아교육연구, 8(3), pp.21~53 (2007), 교육연구소[DOI]
  5. 이진령, "대학원생 핵심역량 검사도구 개발 - P 대학 사례를 중심으로", 교육혁신연구, 30(2), pp.349~371 (2020), 교육발전연구소[DOI]
  6. 이진숙, "대학의 AI 기반 맞춤형 강의 추천 시스템 개발 및 실제 적용 사례 연구: K 대학을 중심으로", 교육공학연구, 37(2), pp.267~307 (2021), 한국교육공학회[DOI]
  7. 조윤정, "인공지능 기반으로 맞춤 및 적응형 학습 시스템의 고등 교육에서의 적용효과", 정보교육학회논문지, 26(4), pp.249~263 (2022), 한국정보교육학회
  8. Alshammari, A, "Twitter User Modeling Based on Indirect Explicit Relationships for Personalized Recommendations", Computational Collective Intelligence. ICCCI 2019. Lecture Notes in Computer Science, 11683, pp.93~105 (2019), Springer
  9. Beel, J., "Research-paper recommender systems : a literature survey", International Journal on Digital Libraries, 17, pp.305~338 (2016)
  10. Brusilovsky, P, "Authoring Tools for Advanced Technology Learning Environments", pp.377~409 (2003), Springer
  11. Chaturapruek, S., "Studying Undergraduate Course Consideration at Scale", AERA Open, 7(10) (2021)
  12. Collins, A., "Document Embeddings vs. Keyphrases vs. Terms for Recommender Systems: A Large-Scale Online Evaluation", 2019 ACM, 2019, 130~133
  13. Dalipi, F., "MOOC Dropout Prediction Using Machine Learning Techniques:Review and Research Challenges", 2018 IEEE global engineering education conference, pp.1007~1014 (2018)
  14. Goel, Y., "On the Effectiveness of Self-Training in MOOC Dropout Prediction", Open Computer Science, 10(1), pp.246~258 (2020)
  15. Guo, G., "Leveraging Title-Abstract Attentive Semantics for Paper Recommendation", Proceedings of the AAAI Conference on Artificial Intelligence, pp.67~74 (2020)
  16. Kong, X., "VOPRec : Vector Representation Learning of Papers with Text Information and Structural Identity for Recommendation", IEEE Transactions on Emerging Topics in Computing, 9(1), pp.226~237 (2018)
  17. Kreutz, C. K., "Scientific paper recommendation systems : a literature review of recent publications", International Journal on Digital Libraries, 23(4), pp.335~369 (2022)
  18. Le, M., "The Impact of Recommenders on Scientific Article Discovery: The Case of Mendeley Suggest", Proceedings of 1st Workshop on the Impact of Recommender Systems, pp.1~5 (2019)
  19. Lee, J., "Personalized Academic Research Paper Recommendation System", ~ (2013)
  20. Li, J., "A comparative study of keyword extraction algorithms for English texts", Journal of Intelligent Systems, 30(1), pp.808~815 (2021)
  21. Li, X., "Personalised Reranking of Paper Recommendations Using Paper Content and User Behavior", ACM Transactions on Information Systems(TOIS), 37(3), pp.1~23 (2019)
  22. Ma, X., "Personalized Scientific Paper Recommendation Based on Heterogeneous Graph Representation", IEEE Access, 7, pp.79887~79894 (2019)
  23. Maghsudi, S., "Personalized Education in the Artificial Intelligence Era : What to expect next", IEEE Signal Processing Magazine, 38(3), pp.37~50 (2021)
  24. Nishioka, C., "Influence of tweets and diversification on serendipitous research paper recommender systems", PeerJ Computer Science, 6, pp.e273~ (2020)
  25. Page, L. C., "How an artificially intelligent virtual assistant helps students navigate the road to college", Aera Open, 3(4) (2017)
  26. Pardos, Z. A., "Connectionist recommendation in the wild : on the utility and scrutability of neural networks for personalized course guidance", User Modeling and User-Adapted Interaction, 29, pp.487~525 (2019)
  27. Pardos, Z. A., "Designing for serendipity in a university course recommendation system", Proceedings of the tenth international conference on learning analytics & knowledge, pp.350~359 (2020)
  28. Rahdari, B., "User-Controlled Hybrid Recommendation for Academic Papers", Proceedings of the 24th International Conference on Intelligent User Interfaces: Companion, pp.99~100 (2019)
  29. Renuka, S., "An Unsupervised Content-Based Article Recommendation System Using Natural Language Processing", Data Intelligence and Cognitive Informatics: Proceedings of ICDICI 2020, 2020, pp.165~180 (2021)
  30. Sakib, N., "A Collaborative Approach Toward Scientific Paper Recommendation Using Citation Context", IEEE Access, 8, pp.51246~51255 (2020)
  31. Silveira, T., "How good your recommender system is? A survey on evaluations in recommendation", International Journal of Machine Learning and Cybernetics, 10, pp.813~831 (2019)
  32. Sun, J., "A hybrid approach for article recommendation in research social networks", Journal of Information Science, 44(5), pp.696~711 (2018)
  33. Tao, M, "Paper Recommend Based on LDA and PageRank", Artificial Intelligence and Security. ICAIS 2020. Communications in Computer and Information Science, 1254, pp.571~584 (2020), Springer
  34. "How Universities Can Use AI Chatbots to Connect Students and Drive Success" (2023)[URL]
  35. Waheed, W., "A Hybrid Approach Toward Research Paper Recommendation Using Centrality Measures and Author Ranking", IEEE Access, 7, pp.33145~33158 (2019)
  36. Wannapiroon, P., "Development of research-based blended learning model to enhance graduate students’research competency and critical thinking skills", Procedia-Social and Behavioral Sciences, 136, pp.486~490 (2014)
  37. Yao, L., "Research on news keyword extraction technology based on TF-IDF and TextRank", 2019 IEEE, 2019, 452~455
  38. Zhao, X., "A Hybrid Model Based on LFM and BiGRU Toward Research Paper Recommendation", IEEE Access, 8, pp.188628~188640 (2020)

관련 논문AI추천

이 간행물의 다른 논문 (7건)