DOI QR코드

DOI QR Code

페이스북의 '좋아요' 리스트를 이용해 다중 공통 관심사항을 추출하는 기법

Extraction Method of Multi-User's Common Interests Using Facebook's 'like' List

  • 임연주 (한국외국어대학교 정보통신공학과) ;
  • 박상원 (한국외국어대학교 정보통신공학과)
  • 투고 : 2014.11.05
  • 심사 : 2015.01.30
  • 발행 : 2015.06.30

초록

최근 스마트폰 발달로 인터넷 접근이 쉬워짐에 따라 소셜 네트워크 서비스(SNS)의 이용이 손쉬워졌다. 하지만 현재 SNS는 개인의 일상 또는 관심사 공유에 그치며 여러 사용자 간의 공통관심사 파악은 어렵다. 본 논문에서는 SNS를 통해 개인이 아닌 여러 사용자 간의 공통관심사를 파악하여 스마트폰을 통해 원하는 것을 추천해주는 콘텐츠 추천 시스템을 제안한다. 추천 시스템은 그룹 내 사용자들의 선호도와 편차를 고려하여 제안한 공식을 포함한다. 시뮬레이션 후 공식에 대해 나올 수 있는 경우는 4가지로 간추려졌다. 그 결과 개인의 선호도를 나타내는 '좋아요' 수가 많으면서 페이스북 사용자들 간 선호도 편차가 적은 콘텐츠를 추천한다. 제안한 방법은 공식에 대한 4가지 경우의 시뮬레이션과 실제 페이스북 사용자들의 '좋아요' 데이터로 증명한다. 제안 시스템은 그룹 내에서의 선호도와 편차를 고려하여 공통관심사를 추천해주기 때문에 양질의 맞춤형 콘텐츠를 제공한다.

The today's rapid spread of smartphones makes it easier to use SNS. However, it reveals only their daily life or interest. Therefore, it is hard to really get to know the detailed part of multi-user's common interests. This paper proposes a content recommendation system which recommends people wanted by identifying common interests through SNS. Recommendation system includes proposal formula considering people wanted and deviation in group. After simulation, the proposed system provide high-quality adapted contents to many users by recommendation item according to the common interest. Number of cases about formula are four. It recommend contents that they have many number of 'like' and few number of deviation in users. The proposed system proves by simulations of four cases and read user's 'likes' data. It provide high-quality adapted contents to many users by recommendation item according to the common interest.

키워드

참고문헌

  1. Park, H. J., Rho, S. K., "Finding Influential Users in the SNS Using Interaction Concept: Focusing on the Blogosphere with Continuous Referencing Relationships," The Journal of Society for e-Business Studies, Vol.17, No.4, pp.69-93, 2012. https://doi.org/10.7838/jsebs.2012.17.4.069
  2. Jong-Gook Bae, Jae-Dong Yang, and Ho-Sang Jo, "Expert Recommending System with Extended Object-Based Thesauri(XOT) for Social Network Service," Journal of Korean Institute of Information Scientists and Engineers, Vol.39, No.6, pp.473-487, 2012.
  3. Lee, H. G., "The exploratory study of the meaning of Facebook : Users' cognition of communication context and the mode of communication," Journal of Cyber Communication Academic Society, Vol.28, No.4, pp.129-172, 2011.
  4. Myong-Ok Kim, Mi-Sun Lee, "A Study on the Effect of Communication Using Facebook on Organization Culture and Emotional Labor: Focusing on K QuasiNon-Governmental Organization," Journal of the Society for e-Business Studies, Vol.18, No.2, pp.131-152, 2013. https://doi.org/10.7838/jsebs.2013.18.2.131
  5. Gyu-Dong Bak, Seung-Jae Oh, Bo-Ra Gang, and Hyo-Jeong So, "Synchronous Feedback Mechanism in Online Social Rating Systems: Examining Social Interaction Effect," Journal of the HCI Society of Korea, Vol.2014, No.2, pp.785-788, 2014.
  6. Saim Shin, Sung-Ju Park, Se-Jin Jang, and Seok-Pil Lee, "Developing the contents recommendation agent using the social user clusters in the social network services on the multiple media convergence environment," Journal of korea information processing society, Vol.38, No.2, 2011.
  7. Seo-Hun Jean, Gi-Hwan Gim, and Jeon-Su Jeon, "Study on Personalized Recommendation Optimization Plan," korea university, 2013.
  8. James Davidson, Benjamin Liebald, and Junning Liu, "The YouTube Video Recommendation System," RecSys, "10Proceedings of the fourth ACM conference on Recommender system," pp.293-296, 2010.
  9. SeokJong Yu, "The dynamic competitive recommendation algorithm in social network services," Information Sciences, pp. 1-14. 2012.
  10. Yohan Jin, Minqing Hu, Harbir Singh, Daniel Rule, and Mikhail Berlyant, "Myspace Video Recommendation with Map-Reduce on Qizmt," Semantic Computing(ICSC), IEEE Fourth International Conference on Semantic Computing, pp.126-133, 2010.
  11. FacebookDeveloper [Internet], https://developers.facebook.com/docs/ plugins/, 2014.
  12. Simplify360, "1 million Facebook fans brings in an average of 826 likes and 309 comments per post," [Internet] http://thenextweb.com/facebook/2011/05/17/1-million-facebook-fansbrings-in-an-average-of-826-likes-and-309-comments-perpost/
  13. Je Hyok Rew, Young Hwan Choi, and Een Jun Hwang, "Human Computer Interaction: A Facebook Page Ranking and Highlight Contents Selection Scheme based on User Interests," korea information processing society, Vol.3, No.2, pp.101-108, 2014.