$\require{mediawiki-texvc}$

연합인증

연합인증 가입 기관의 연구자들은 소속기관의 인증정보(ID와 암호)를 이용해 다른 대학, 연구기관, 서비스 공급자의 다양한 온라인 자원과 연구 데이터를 이용할 수 있습니다.

이는 여행자가 자국에서 발행 받은 여권으로 세계 각국을 자유롭게 여행할 수 있는 것과 같습니다.

연합인증으로 이용이 가능한 서비스는 NTIS, DataON, Edison, Kafe, Webinar 등이 있습니다.

한번의 인증절차만으로 연합인증 가입 서비스에 추가 로그인 없이 이용이 가능합니다.

다만, 연합인증을 위해서는 최초 1회만 인증 절차가 필요합니다. (회원이 아닐 경우 회원 가입이 필요합니다.)

연합인증 절차는 다음과 같습니다.

최초이용시에는
ScienceON에 로그인 → 연합인증 서비스 접속 → 로그인 (본인 확인 또는 회원가입) → 서비스 이용

그 이후에는
ScienceON 로그인 → 연합인증 서비스 접속 → 서비스 이용

연합인증을 활용하시면 KISTI가 제공하는 다양한 서비스를 편리하게 이용하실 수 있습니다.

네트워크 분석을 활용한 딥러닝 기반 전공과목 추천 시스템
Major Class Recommendation System based on Deep learning using Network Analysis 원문보기

지능정보연구 = Journal of intelligence and information systems, v.27 no.3, 2021년, pp.95 - 112  

이재규 (연세대학교 산업공학과) ,  박희성 (연세대학교 산업공학과) ,  김우주 (연세대학교 산업공학과)

초록
AI-Helper 아이콘AI-Helper

대학 교육에 있어서 전공과목의 선택은 학생들의 진로에 중요한 역할을 한다. 하지만, 산업의 변화에 발맞춰 대학 교육도 학과별 전공과목의 분야가 다양해지고 그 수가 많아지고 있다. 이에 학생들은 본인의 진로에 맞게 수업을 선택하여 수강하는 것에 어려움을 겪고 있다. 본 연구는 대학 전공과목 추천 모델을 제시함으로써 개인 맞춤형 교육을 실현하고 학생들의 교육만족도를 제고하고자 한다. 모델 연구에는 대학교 학부생들의 2015년~2017년 수강 이력 데이터를 활용하였으며, 메타데이터로는 학생과 수업의 전공 명을 사용했다. 수강 이력 데이터는 컨텐츠 소비 여부만을 나타낸 암시적 피드백 데이터로, 수업에 대한 선호도를 반영한 것이 아니다. 따라서 학생과 수업의 특성을 나타내는 임베딩 벡터를 도출했을 시, 표현력이 낮다. 본 연구는 이러한 문제점에 착안하여, 네트워크 분석을 통해 학생, 수업의 벡터를 생성하고 이를 모델의 입력 값으로 활용하는 Net-NeuMF 모델을 제시한다. 모델은 암시적 피드백을 가진 데이터를 이용한 대표적인 모델인 원핫 벡터를 이용하는 NeuMF의 구조를 기반으로 하였다. 모델의 입력 벡터는 네트워크 분석을 통해 학생과 수업의 특성을 나타낼 수 있도록 생성하였다. 학생을 표현하는 벡터를 생성하기 위해, 각 학생을 노드로 설정하고 엣지는 두 학생이 같은 수업을 수강한 경우 가중치를 가지고 연결되도록 설계했다. 마찬가지로 수업을 표현하는 벡터를 생성하기 위해 각 수업을 노드로 설정하고 엣지는 공통으로 수강한 학생이 있는 경우 연결시켰다. 이에 각 노드의 특성을 수치화 하는 표현 학습방법론인 Node2Vec을 이용하였다. 모델의 평가를 위해 추천 시스템에서 주로 활용하는 지표 4가지를 사용하였고, 임베딩 차원이 모델에 미치는 영향을 분석하기 위해 3가지 다른 차원에 대한 실험을 진행하였다. 그 결과 기존 NeuMF 구조에서 원-핫 벡터를 이용하였을 때보다 차원과 관계없이 평가지표에서 좋은 성능을 보였다. 이에 본 연구는 학생(사용자)와 수업(아이템)의 네트워크를 이용해 기존 원-핫 임베딩 보다 표현력을 높였다는 점, 모델을 구성하는 각 구조의 특성에 맞도록 임베딩 벡터를 활용하였다는 점, 그리고 기존의 방법론에 비해 다양한 종류의 평가지표에서 좋은 성능을 보였다는 점을 기여점으로 가지고 있다.

Abstract AI-Helper 아이콘AI-Helper

In university education, the choice of major class plays an important role in students' careers. However, in line with the changes in the industry, the fields of major subjects by department are diversifying and increasing in number in university education. As a result, students have difficulty to c...

주제어

표/그림 (11)

참고문헌 (20)

  1. Dudani, Sahibsingh A. "The distance-weighted k-nearest-neighbor rule." IEEE Transactions on Systems, Man, and Cybernetics 4 (1976): 325-327. 

  2. Grover, Aditya, and Jure Leskovec. "node2vec: Scalable feature learning for networks." Proceedings of the 22nd ACM SIGKDD international conference on Knowledge discovery and data mining. (2016) 

  3. Han. Ji Won, HS. Lim, "The Recommendation System based on Collaborative Filtering for Adaptive Learning", The Korean Association Of Computer Education a collection of essays 157-160, (2015) 

  4. Harper, F. Maxwell, and Joseph A. Konstan. "The movielens datasets: History and context.", Acm transactions on interactive intelligent systems (tiis) 5.4 1-19 (2015) 

  5. He, Xiangnan, et al., "Neural collaborative filtering.", Proceedings of the 26th international conference on world wide web. (2017) 

  6. Herlocker, Jonathan L., Joseph A. Konstan, and John Riedl. "Explaining collaborative filtering recommendations.", Proceedings of the 2000 ACM conference on Computer supported cooperative work. (2000) 

  7. Kim Du hyeung, WS. Shin, KW. Han, JS. Lee, KB. Moon, SG. Lee, SY. Han, HJ. Kwon, SW. Han, "A system for recommending university liberal arts courses using collaborative filtering", Korean Institute Of Industrial Engineers Autumn Conference a collection of essays 2551-2556, (2020) 

  8. Kim Eun-Seok, "A Study on the Difference of Job Satisfaction by Means of Major Selection Criteria of College Graduates", The Journal of Career Education Research 28(3) 85-101, (2015) 

  9. Kim Gui-Jung, BH. Kim, JS. Han, "Customizing Intelligent Recommendation System based on Compound Knowledge", JOURNAL OF THE KOREA CONTENTS ASSOCIATION 10(8) 26-31, (2010) 

  10. Koo Yoo Young, DH Park, JJ Kim, YH Park, CK Ko, BK Lee, "Meta-analysis of course selection data of the university graduates revealed the problems of course structures", Korean Journal of General Education 13(2) 369-396, (2019) 

  11. Lee Hayeon, JE. Go, MH. Joo, "Effects of University belonging and College Life Satisfaction on Learning Persistence in Non-face-to-face Learning Environment due to COVID-19 Pandemic", The Journal of Career Education Research 34(1) 231-251, (2021) 

  12. Mikolov, Tomas, et al., "Efficient estimation of word representations in vector space.", arXiv preprint arXiv:1301.3781 (2013). 

  13. P. D. Hoff, A. E. Raftery, and M. S. Handcock. "Latent space approaches to social network analysis", J. of the American Statistical Association, (2002) 

  14. Park Roh-Gook, "A Study on Selection Altenatives of Basic and Major Courses for College Students", Journal of the Korea Industrial Information Systems Research 6(1) 48-55, (2001) 

  15. S. Fortunato. "Community detection in graphs", Physics Reports, 486(3-5):75 - 174, 2010. 

  16. Sarwar, Badrul, et al., "Item-based collaborative filtering recommendation algorithms.", Proceedings of the 10th international conference on World Wide Web. (2001) 

  17. Wang Xiang, H. Xiangnan, Y. Cao, M. Liu and T. Chua, "KGAT: Knowledge Graph Attention Network for Recommendation", The 25th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD '19), August 4-8, (2019) 

  18. Y. Zheng, B. Tang, W. Ding, and H. Zhou. "A neural autoregressive approach to collaborative filtering", In ICML, 764-773, (2016) 

  19. Yehuda Koren, R. Bell, C. Volinsky, "Matrix factorization techniques for recommender systems.", IEEE Computer 42(8) 30-37, (2009) 

  20. Zhang, Fuzheng, et al., "Collaborative knowledge base embedding for recommender systems.", Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining. (2016). 

저자의 다른 논문 :

관련 콘텐츠

오픈액세스(OA) 유형

GOLD

오픈액세스 학술지에 출판된 논문

이 논문과 함께 이용한 콘텐츠

저작권 관리 안내
섹션별 컨텐츠 바로가기

AI-Helper ※ AI-Helper는 오픈소스 모델을 사용합니다.

AI-Helper 아이콘
AI-Helper
안녕하세요, AI-Helper입니다. 좌측 "선택된 텍스트"에서 텍스트를 선택하여 요약, 번역, 용어설명을 실행하세요.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.

선택된 텍스트

맨위로