• Journal of the Korean Society for information Management
• /
• v.8 no.1
• /
• pp.100-116
• /
• 1991

In this study, a retrieval system taking advantage of term relationship in a specific domain and also of evidential reasoning as tools for measuring relevance is implemented. For this experiment, techincal memoranda documented in Electronics and Telecommunications Research Institute (ETRI) served as a sample document file. Sample knowledge base was prepared by extracting terms and term relations pertaining to telecommunications from INSPEC thesaurus. Relations between terms were represented by numerical values according to types of term relations. Relationship between a query and a document was measured according to Dempster-Shafer theory of evidence. As a result of this experiment, a more comprehensive search was made by expanding search terms utilizing term relations. Measure of relevance represented by reflecting term relations, and search results were listed in a descending order of relevance.

• Journal of Korean Library and Information Science Society
• /
• v.31 no.4
• /
• pp.139-164
• /
• 2000

본 연구는 이용자 지향적 평가의 의의와 타당성을 규명하기 위하여 연구과정에서 개인의 지식과 상황 및 인지적 요인의 변화에 따라 이용자가 인식하는 검색문헌의 적합성을 평가하는 기준들의 변화를 규명하는데 그 목적이 있다. 연구의 방법은 이론가설을 설정하고 이의 검증을 위하여 실제 연구문제를 가지고 있는 5명의 이용자를 대상으로 그들의 연구과정에 따라 적합성을 평가하는 기준들과 그 기준들의 변화양상을 관찰하는 실험을 수행하였다. 실험결과의 분석에는 빈도분석, 상관분석, 일원분산분석이 활용되었다. 그 결과 연구자들은 연구과정에 따라 적합성을 평가하는 기준이 다르게 나타났으며, 평가자에 따라서도 각기 다른 평가기준이 사용되고 있음을 확인할 수 있었다.

• Asia pacific journal of information systems
• /
• v.15 no.3
• /
• pp.223-241
• /
• 2005

In spite of the rapid growth of mobile multimedia contents market, most of the customers experience inconvenience, lengthy search processes and frustration in searching for the specific multimedia contents they want. These difficulties are attributable to the current mobile Internet service method based on inefficient sequential search. To overcome these difficulties, this paper proposes a MOBIIe COntents Recommender System for Movie(MOBICORS-Movie), which is designed to reduce customers' search efforts in finding desired movies on the mobile Internet. MOBICORS-Movie consists of three agents: CF(Collaborative Filtering), CBIR(Content-Based Information Retrieval) and RF(Relevance Feedback). These agents collaborate each other to support a customer in finding a desired movie by generating personalized recommendations of movies. To verify the performance of MOBICORS-Movie, the simulation-based experiments were conducted. The results from this experiments show that MOBICORS-Movie significantly reduces the customer's search effort and can be a realistic solution for movie recommendation in the mobile Internet environment.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.43 no.1 s.307
• /
• pp.53-66
• /
• 2006

This paper proposes the combined image retrieval system that gives the same relevance as exhaustive search method while its performance can be considerably improved. This system is combined with two different retrieval methods and each gives the same results that full exhaustive search method does. Both of them are two-stage method. One uses condensation of feature vectors, and the other uses binary-tree clustering. These two methods extract the candidate images that always include correct answers at the first stage, and then filter out the incorrect images at the second stage. Inasmuch as these methods use equal algorithm, they can get the same result as full exhaustive search. The first method condenses the dimension of feature vectors, and it uses these condensed feature vectors to compute similarity of query and images in database. It can be found that there is an optimal condensation ratio which minimizes the overall retrieval time. The optimal ratio is applied to first stage of this method. Binary-tree clustering method, searching with recursive 2-means clustering, classifies each cluster dynamically with the same radius. For preserving relevance, its range of query has to be compensated at first stage. After candidate clusters were selected, final results are retrieved by computing similarities again at second stage. The proposed method is combined with above two methods. Because they are not dependent on each other, combined retrieval system can make a remarkable progress in performance.

• Journal of the Korean Society for Library and Information Science
• /
• v.27
• /
• pp.265-289
• /
• 1994

Databases publicly available for online searching permit both citation and subject searching, however, subject searching has dominated the online search environment. Despite the power of citation searching, it may be underutilized This study explored the relationship between the number of cited references used in a citation search and information retrieval effectiveness, a relatively unstudied phenomenon. Three articles in the library and information science literature were chosen to represent sample questions. Cited reference searches were conducted for each article and each of its references. All searches were conducted in Social Scisearch and Scisearch on DIALOG. Relevance judgments on the retrieved citations were obtained from the authors of the original articles. This research focused on analyzing, in terms of information retrieval effectiveness, the overlap among postings sets retrieved by various combinations of cited references. The findings from the three case studies clearly showed that the more cited references used for the citation search, the better the performance, in terms of retrieving more relevant documents, up to a point of diminishing retums. In addition, generally the overall level of overlap among relevant documents sets was found to be low. Therefore, if only some of the cited references among many candidates are used for a citation search, a significant proportion of relevant documents may be missed. The analysis of the characteristics of cited references provided the ways to predict which cited refereces would be useful to improve information retrieval. The findings of this comprehensive exploratory study are of interest for both theoretical and practical reasons. They contribute to the development of a theoretical model for the effective use of the citation search. This model might also be implemented in operational online systems. In addition, the findings potentially will help online searchers improve their search strategies using the citation search so that they can better achieve their information retrieval goals: the retrieval of items relevant to a given question and the suppression of nonrelevant items.

• Journal of KIISE:Databases
• /
• v.33 no.4
• /
• pp.372-383
• /
• 2006

Among the approaches to image database management, content-based image retrieval (CBIR) is viewed as having the best support for effective searching and browsing of large digital image libraries. Typical CBIR systems allow a user to provide a query image, from which low-level features are extracted and used to find 'similar' images in a database. However, there exists the semantic gap between human visual perception and low-level representations. An effective methodology for overcoming this semantic gap involves relevance feedback to perform feature re-weighting. Current approaches to feature re-weighting require the number of components for a feature representation to be the same for every image in consideration. Following this assumption, they map each component to an axis in the n-dimensional space, which we call the metric space; likewise the feature representation is stored in a fixed-length vector. However, with the emergence of features that do not have a fixed number of components in their representation, existing feature re-weighting approaches are invalidated. In this paper we propose a feature re-weighting technique that supports features regardless of whether or not they can be mapped into a metric space. Our approach analyses the feature distances calculated between the query image and the images in the database. Two-sided confidence intervals are used with the distances to obtain the information for feature re-weighting. There is no restriction on how the distances are calculated for each feature. This provides freedom for how feature representations are structured, i.e. there is no requirement for features to be represented in fixed-length vectors or metric space. Our experimental results show the effectiveness of our approach and in a comparison with other work, we can see how it outperforms previous work.

• Science of Emotion and Sensibility
• /
• v.15 no.3
• /
• pp.367-380
• /
• 2012

The purpose of this research was to implement and verify an information retrieval(IR) system based on users' relevance criteria for information search tasks. For this purpose, we implemented an IR system with a dynamic ranking model using users' relevance criteria varying with the types of information search task and evaluated this system through user experiment. 45 participants performed three information search tasks on both IR systems with a static and a dynamic ranking model. Three Information search tasks are fact finding search task, problem solving search task and decision making search task. Participants evaluated top five search results on 7 likert scales of relevance. We observed that the IR system with a dynamic ranking model provided more relevant search results compared to the system with a static ranking model. This research has significance in designing IR system for information search tasks, in testing the validity of user-oriented relevance judgement model by implementing an IR system for actual information search tasks and in relating user research to the improvement of an IR system.

• Annual Conference on Human and Language Technology
• /
• 1997.10a
• /
• pp.227-232
• /
• 1997

이 논문은 인터넷 홈페이지를 검색하는 정보검색시스템인 미리내 시스템을 제시한다. 웹 문서의 특성을 고려하여 로봇의 기능을 확장하고, 색인, 등록, 수정, 삭제, 분류의 자동화를 구현하여 관리효율을 높인다. 자동화에 따른 문제점과 해결방법을 제시하고, 불리언질의검색 외에 자연언어질의 검색에서 질의어 확장의 방법으로 웹페이지 링크속성검색, Relevance feedback을 통한 검색효율을 높인다.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.23 no.4
• /
• pp.141-149
• /
• 1998

One of the difficulties in using the current Boolean-based information retrieval systems is that it is hard for a user, especially a novice, to formulate an effective Boolean query. One solution to this problem is to let the system formulate a query for a user from his relevance feedback documents in this research, an intelligent query reformulation mechanism based on ID3 is proposed and the sensitivity of its retrieval effectiveness, i.e., recall, precision, and E-measure, to various input settings is analyzed. The parameters in the input settings is the number of relevant documents. Experiments conducted on the test set of Medlars revealed that the effectiveness of the proposed system is in fact sensitive to the number of the initial relevant documents. The case with two or more initial relevant documents outperformed the case with one initial relevant document with statistical significances. It is our conclusion that formulation of an effective query in the proposed system requires at least two relevant documents in its initial input set.

• Journal of the Korean Society for information Management
• /
• v.8 no.2
• /
• pp.96-107
• /
• 1991

The inherent limitation of the conventional approaches in automatic indexing lies in the fact that they compute the relevancy between index terms and documents rather indirectly or relatively. As an alternative the anlaysis of document texts seeks a means of establishing a direct relevancy of the terms. More rigorous linguistic analysis will ensure better chance of relevancy. Various semantic topologies among terms may suggest the sufficient quality for relevancy. The enhanced and guaranteed relevance will allow the high precision of retrieval. Along with this line the on going project in KAIST pursues the user oriented retrieval system that spawns still may other issues that are not c o m n in traditional perspective.