Journal of KIISE:Information Networking (한국정보과학회논문지:정보통신)

Korean Institute of Information Scientists and Engineers (한국정보과학회)
권호 표지

Globally Optimal Recommender Group Formation and Maintenance Algorithm using the Fitness Function

적합도 함수를 이용한 최적의 추천자 그룹 생성 및 유지 알고리즘

  • 김용구 (삼성전자 디지털미디어연구소) ;
  • 이민호 (삼성전자 디지털미디어연구소) ;
  • 박수홍 (삼성전자 디지털미디어연구소) ;
  • 황철주 (삼성전자 디지털미디어연구소)
  • Published : 2009.02.15
Download PDF
⟨ Previous Next ⟩

Abstract

This paper proposes a new algorithm of clustering similar nodes defined as nodes having similar characteristic values in pure P2P environment. To compare similarity between nodes, we introduce a fitness function whose return value depends only on the two nodes' characteristic values. The higher the return value is, the more similar the two nodes are. We propose a GORGFM algorithm newly in conjunction with the fitness function to recommend and exchange nodes' characteristic values for an interest group formation and maintenance. With the GORGFM algorithm, the interest groups are formed dynamically based on the similarity of users, and all nodes will highly satisfy with the information recommended and received from nodes of the interest group. To evaluate of performance of the GORGFM algorithm, we simulated a matching rate by the total number of nodes of network and the number of iterations of the algorithm to find similar nodes accurately. The result shows that the matching rate is highly accurate. The GORGFM algorithm proposed in this paper is highly flexible to be applied for any searching system on the web.

본 논문에서는 P2P 네트워크 환경에서 유사한 특성을 가진 다른 노드(node)를 찾아 추천자(recommender) 그룹을 형성하고 유지하는 새로운 알고리즘을 제안한다. 두 노드의 유사한 특성을 비교하기 위해 본 논문에서는 두 노드의 특성값(characteristic value. 이하 CV)을 이용한 적합도 검사(fitness evaluation)를 사용하여 유사도(similarity)를 확인한다. 유사도의 크기가 작을수록 두 노드는 매우 유사한 특성을 가지게 된다. 또한, 본 논문에서 제안하는 GORGFM(Globally Optimal Recommender Group Formation and Maintenance) 알고리즘은 최단 기간 내에 최적의 추천자 그룹을 형성하고 사용자의 선호도 변화에 대응할 수 있는 알고리즘이다. GORGFM 알고리즘을 평가하기 위해 본 논문에서는 매칭율(matching rate)과 얼마나 빠르고 정확하게 추천자 그룹을 형성하는가에 대해 시뮬레이션 한다. GORGFM 알고리즘은 네트워크에서뿐만 아니라 인터넷상에서 컨텐츠(contents) 검색 등과 같이 적합도 함수(fitness function)를 이용할 수 있는 모든 시스템에 적용할 수 있다.

Keywords

References

  1. Goldberg, D.E., Genetic algorithms in search,optimization, and machine learning, Reading, MA,Addison-Wesley, 1989
  2. Adeli, H. and Hung, S., Machine learning : neural networks, genetic algorithms, and fuzzy systems,New York, Wiley, 1995
  3. Babu, G.P. and Murty, M.N., "A near-optimalinitial seed value selection in K--Means algorithmusing a genetic algorithm," Pattern RecognitionLetters, Vo1.l4, No.10, 1993, pp. 763-769 https://doi.org/10.1016/0167-8655(93)90058-L
  4. Murthy, c.A. and Chowdhury, N., "In search ofoptimal clusters using genetic algorithms," PatternRecognition Letters, Vo17, 1996, pp. 825-832 https://doi.org/10.1016/0167-8655(96)00043-8
  5. Sexton, RS., Dorsey, RE., and Johnson, J.D.,"Toward global optimization of neural networks :A comparison of the genetic algorithm and backpropagation," Decision Support Systems, Vol.22,1998, pp. 171-185 https://doi.org/10.1016/S0167-9236(97)00040-7
  6. Pena, J,M., Lozano, J,A., and Larranaga, P., "Anempirical comparison of four initialization methodsfor the K-Means algorithm," Pattern RecognitionLetters, Vol.20, 1999, pp. 1027-1040 https://doi.org/10.1016/S0167-8655(99)00069-0
  7. Sexton, RS., Dorsey, RE., and Johnson, J.D.,"Optimization of neural networks : A comparativeanalysis of the genetic algorithm and simulatedannealing," European Journal of Operational Research,Vo1.114, 1999, pp. 589-601 https://doi.org/10.1016/S0377-2217(98)00114-3
  8. Fuketa, M., Morita, K., Shishibori, M., lun IchiAoe, "Intelligent system of selecting key searchalgorithms automatically," Systems, Man, andCybernetics, 1996., IEEE International Conferenceon Volume 1, pp. 642-647
  9. Bin He, Yan Qiu, Ze·ying Li, "Similarity Reasoningand Similarity Transfomlations," Controland Automation, 2007. ICCA 2007. IEEE InternationalConference on 2007. pp. 1667-1671