바로가기메뉴

본문 바로가기 주메뉴 바로가기

logo

  • P-ISSN1013-0799
  • E-ISSN2586-2073

기계학습 기반 피드백 과정을 통한 SDI 시스템의 성능향상에 관한 연구

Machine-Learning Based on Relevance Feedback: A Powerful Engine to Enhance the Performance of SDI System

정보관리학회지, (P)1013-0799; (E)2586-2073
2004, v.21 no.4, pp.133-152
https://doi.org/10.3743/KOSIM.2004.21.4.133
노영희 (건국대학교)

  • 다운로드 수
  • 조회수

초록

정보시대의 도래로 정보량은 기하급수적으로 증가하게 되었고, 이러한 대량의 정보로부터 이용자 개개인에게 적합한 정보를 적시에 제공할 수 있는 방법으로 SDI 서비스가 연구개발되어 왔지만, 현실적으로 그 활용도는 매우 낮은 것으로 조사되었다. 이에 본 논문에서는 그 원인을 분석하고 SDI 시스템의 성능을 개선시킬 수 있는 적합성 피드백 기반 SDI 시스템을 개발하고자 하였다. 본 연구의 실험을 위해 개발된 실험시스템은 이용자 최소개입 피드백기반 SDI 시스템, 완전자동 피드백기반 SDI 시스템, 그리고 이용자 최대개입 피드백 기반 SDI 시스템이며, 새로 개발된 3개 시스템의 성능 개선정도를 평가하기 위해 네 번째 시스템으로서 전통적인 SDI 서비스에서 사용하고 있는 방법으로 시스템을 개발하였다. 실험결과 이용자 최대개입 피드백 기반 SDI 시스템이 가장 높은 성능을 보여 주었고, 완전자동 피드백 기반, 이용자 최소개입 피드백기반, 전통적 SDI 시스템 순으로 나타났으며, 피드백 기반 시스템들은 피드백이 진행될수록 그 성능이 향상되는 것으로 나타났다.

Abstract

As the Internet facilitates the rapid increase of information availability, the study on SDI service that provides users with relevant document in a timely manner has been developed. However, the practical use of this service has been low. This thesis aims at analyzing the reasons for this and developing relevance feedback based SDI system to improve the performance of the existing SDI system. Experimental systems that are developed for this study are SDI system based on users' minimum intervention feedback, SDI system based on perfect automation feedback, and SDI system based on users' maximum intervention feedback. The fourth system that utilizes the traditional SDI system is also studied to evaluate the level of performance improvement of the newly developed three types of SDI system. As a result of this study, SDI system based on users' maximum intervention feedback showed greatest performance improvement. The next performance improvement happened in order of SDI system based on perfect automation feedback, SDI system based on users' minimum intervention feedback, and the traditional SDI system. Feedback based systems showed greater performance improvement as they went through more feedback processes.

참고문헌

1

노영희. (2003). 국내 대학도서관의 SDI 서비스 제공현황 분석 및 통합형 서비스 시스템 구축 방안에 관한 연구. 정보관리학회지, 20(3), 199-203.

2

정영미. (1993). 정보검색론:구미무역.

3

Amati, Gianni. (1997). Learning System for Selective Dissemination of Information (764-769). 15th International Joint Conference on Artificial Intelligence.

4

Belkin, N. J. (1992). “Information Filtering and Information Retrieval: Two Sides of the Same Coi. Communications of the ACM,, 35(12), 29-38.

5

Blake, P. (1997). Exploring the News:Information World Review.

6

Bonifati, Angela. (2001). Pushing Reactive Services to XML Repositories using Active Rules. 39(5), 633-641.

7

Boughanem, M. (2002). Incremental Adaptive Filtering: Profile Learning and Threshold Calibration.. ACM, , 640-644.

8

Callan, J. P. (1995). TREC and TIPSTER Experiments with INQUERY. Information Processing and Management, 31(3), 327-343.

9

Callan, Jamie.. (1119). Document Filtering with Inference Networks. SIGIR, 96, 262-269.

10

Chung, Young Mee. (2003). A Study on Automatic text categorization of internet document. Journal of Information Science, 29(1), 117-126.

11

Dasarathy, Belur V. (1991). Nearest Neighbor(NN) Norms: NN Patern Classification Techniuqes. McGraw- Hill Computer Science Series. Las Alamitos,:IEEE Computer Society Press..

12

Fischer, G. (1991). Information Access in Complies, Poorly Structured Information Spaces (63-70). ACM Special Interest Group on Human Computer Interaction Annual Conference.

13

Foltz, P. W. (1992). Personalized Information Delivery: An Analysis of Information Filtering Methods. Communications of the ACM, 35(12), 51-60.

14

Frants, V. I. One Approach to Classification of Users and Automatic Clustering of Documents. Information Processing and Management, 29(2), 187-195.

15

Goker, A. (1991). Toward an Adaptive Information Retrieval System (348-357). 6th International Symposium.

16

Iwayama, Makato. (1995). Cluster-based Text Categorization: A Comparison of Category Search Strategies (273-281). 18th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR).

17

Lang, K. (1995). NewsWeeder: An adaptive multi-user text filter:Tech. Rep., School of Computer Science.

18

Manlone, T. W. (1987). Intelligent Information Sharing Systems. Com- mun. ACM, 30(5), 390-402.

19

Masand, B., G. (1992). “Classifying News Stories Using Memory based Reseonin.

20

Rich, E. (1983). Users are Individuals: Individualizing User Models. International Journal. Man-Mach. Studies, 18, 199-214.

21

Robertson, S. E., S. (1995). Walker, S. Jones, M. M. Hancock-Beaulieu, and M. Gatford. Okapi at TREC-3 (500-225). The Third Text Retrieval Conference (TREC- 3).

22

Salton, Gerald. (1983). Introduction to Modern Information:Retrieval. McGraw-Hill.

23

Seth, B. D. A Learning Approach to Personalized Information Filtering..

24

Singhal, Amit. (1997). “Learning Routing Queries in a Query Zone. SIGIR, 97, 25-32.

25

Walker, S. (1997). :Kkapi/Keebow ant TREC-6 automatic and ad hoc, VLC, Routing, Filtering and QSDR.

26

Wyle, M. F. Retrieving Highly Dynamic Distributed Information. (108-115). ACM SIGIR lnternational Conference on Research and Development in Information Retrieval:.

27

Yan T. (1995). SIFT - A tool for wide-area information dissemination (177-186). 1995 USENIX Technical Conference.

28

Yan T. (1994). Distributed Selective Dissemination of Information (89-98). 3rd International Conference on Parallel and Distributed Information Systems (PDIS, Austin, TX, Sept.).

29

Yang, Y. Expert Network: Effective and Efficient Learning from Human Decisions in Text Categorization and Retrieval (11-21). 17th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR).

30

Yochum, J. A. A High-Speed Text Scanning Algorithm Utilizing Least Frequent Trigraph. (114-121). IEEE International Symposium on New Directions in Computing.

정보관리학회지