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.

초록

Abstract

This study constructed an ontology targeting journal articles and evaluated its performance. Also, the performance of a triple structure ontology was compared with the knowledge base of an inverted index file designed for a simple keyword search engine. The coverage was three years of articles published in the Journal of the Korean Society for Information Management from 2007 to 2009. Protégé was used to construct an ontology, whilst utilizing an inverted index file to compare performance. The concept ontology was manually established, and the bibliography ontology was automatically constructed to produce an OWL concept ontology and an OWL bibliography ontology, respectively. This study compared the performance of the knowledge base of the ontology, using the Jena search engine with the performance of an inverted index file using the Lucene search engine. As a result, The Lucene showed higher precision rate, but Jena showed higher recall rate.