• Journal of KIISE:Databases
• /
• v.34 no.3
• /
• pp.193-205
• /
• 2007

Increasing amount of digital images requires more accurate and faster way of image retrieval. So far, image retrieval method includes content-based retrieval and keyword based retrieval, the former utilizing visual features such as color and brightness and the latter utilizing keywords which describe the image. However, the effectiveness of these methods as to providing the exact images the user wanted has been under question. Hence, many researchers have been working on relevance feedback, a process in which responses from the user are given as a feedback during the retrieval session in order to define user’s need and provide improved result. Yet, the methods which have employed relevance feedback also have drawbacks since several feedbacks are necessary to have appropriate result and the feedback information can not be reused. In this paper, a novel retrieval model has been proposed which annotates an image with a keyword and modifies the confidence level of the keyword in response to the user’s feedback. In the proposed model, not only the images which have received positive feedback but also the other images with the visual features similar to the features used to distinguish the positive image are subjected to confidence modification. This enables modifying large amount of images with only a few feedbacks ultimately leading to faster and more accurate retrieval result. An experiment has been performed to verify the effectiveness of the proposed model and the result has demonstrated rapid increase in recall and precision while receiving the same number of feedbacks.

• Journal of the Korean Society for information Management
• /
• v.32 no.3
• /
• pp.277-293
• /
• 2015

This study examined the relative retrieval effectiveness after relevance feedback between two systems (Title/Abstract and Full-text) using four different sets of relevance judgment. Four relevance levels (not relevant, marginally relevant, relevant, highly relevant) are also used, each of which is determined by referees giving a relevance score to documents. This study also investigated how much the average precision was improved after relevance feedback when "marginally relevant" documents are included in the relevant class with the Title/Abstract system, and with the Full-text retrieval system as well. It is found that the Title/Abstract system benefited from relevance feedback with the marginally relevant documents. In case of the Title/Abstract system, the higher percentage of improvement was consistently obtained when including the marginally relevant documents in the relevance class, however the result was vice versa in case of the Full-text retrieval system. It implied that the marginally relevant documents in the relevant class had caused noises in the Full-text retrieval system.

• Journal of KIISE:Databases
• /
• v.32 no.4
• /
• pp.362-371
• /
• 2005

Similarity indexing and searching are well known to be difficult in high-dimensional applications such as multimedia databases. Especially, they become more difficult when multiple features have to be indexed together. In this paper, we propose a novel indexing method called the GB-index that is designed to efficiently handle complex similarity queries as well as relevance feedback in high-dimensional image databases. In order to provide the flexibility in controlling multiple features and query objects, the GB-index treats each dimension independently The efficiency of the GB-index is realized by specialized bitmap indexing that represents all objects in a database as a set of bitmaps. Main contributions of the GB-index are three-fold: (1) It provides a novel way to index high-dimensional data; (2) It efficiently handles complex similarity queries; and (3) Disjunctive queries driven by relevance feedback are efficiently treated. Empirical results demonstrate that the GB-index achieves great speedups over the sequential scan and the VA-file.

• Journal of the Korea Society of Computer and Information
• /
• v.11 no.4 s.42
• /
• pp.1-7
• /
• 2006

Recently, studies of relevance feedback to increase the performance of image retrieval has been activated. In this Paper a new region weighting method in region based image retrieval with relevance feedback is proposed to reduce the semantic gap between the low level feature representation and the high level concept in a given query image. The new weighting method determines the importance of regions according to the spatial locations of regions in an image. Experimental results demonstrate that the retrieval quality of our method is about 18% in recall better than that of area percentage approach. and about 11% in recall better than that of region frequency weighted by inverse image frequency approach and the retrieval time of our method is a tenth of that of region frequency approach.

• Journal of the Korean Society for information Management
• /
• v.20 no.4 s.50
• /
• pp.23-48
• /
• 2003

In this study an information filtering system was implemented and a series of relevance feedback experiments were conducted using the system. For the relevance feedback, the original queries were searched against the database and the results were reviewed by the researchers. Based on users' online relevance judgements a pair of 17 refined queries were generated using two methods called 'co-occurrence exclusion method' and 'lower frequencies exclusion method,' In order to generate them, the original queries, the descriptors and category codes appeared in either relevant or irrelevant document sets were applied as elements. Users' relevance judgments on the search results of the refined queries were compared and analyzed against those of the original queries.

• The KIPS Transactions:PartB
• /
• v.17B no.2
• /
• pp.99-106
• /
• 2010

In this paper, we propose the construction of multi-queries and consistency principle for the user's emotion-based music retrieval system. The features used in the system are MPEG-7 audio descriptors, which are international standards recommended for content-based audio retrievals. In addition we propose the method to determine the weight that represent the importance of each descriptor for each emotion in order to reduce the computation. Also, the proposed retrieval algorithm that uses the relevance feedback based on consistency principal and multi-queries improves the success ratio of musics corresponding to user's emotion.

• Annual Conference on Human and Language Technology
• /
• 1999.10e
• /
• pp.7-12
• /
• 1999

정보 검색 시스템은 사용자 질의의 키워드들과 문서들의 유사성(similarity)을 기준으로 관련 문서들을 순서화하여 사용자에게 제공한다. 그렇지만 인터넷 검색에 사용되는 질의는 일반적으로 짧기 때문에 보다 유용한 질의를 만들고자 하는 노력이 지금까지 계속되고 있다. 그러나 키워드에 포함된 정보가 제한적이기 때문에 이에 대한 보완책으로 사용자의 적합성 피드백을 이용하는 방법을 널리 사용하고 있다. 본 논문에서는 일반적인 적합성 피드백의 가장 큰 단점인 빈번한 사용자 참여는 지양하고, 시스템에 기반한 적합성 피드백에서 배제한 사용자 참여를 유도하는 검색 문서의 분류 정보에 기반한 용어 클러스터 질의 확장 모델(Term Cluster Query Expansion Model)을 제안한다. 이 방법은 검색 시스템에 의해 검색된 상위 n개의 문서에 대하여 분류기를 이용하여 각각의 문서에 분류 정보를 부여하고, 문서에 부여된 분류 정보를 이용하여 분류 정보의 수(m)만큼으로 문서들을 그룹을 짓는다. 적합성 피드백 알고리즘을 이용하여 m개의 그룹으로부터 각각의 용어 클러스터(Term Cluster)를 생성한다. 이 클러스터가 사용자에게 문서 대신에 피드백의 자료로 제공된다. 실험 결과, 적합성 알고리즘 중 Rocchio방법을 이용할 때 초기 질의보다 나은 성능을 보였지만, 다른 연구에서 보여준 성능 향상은 나타내지 못했다. 그 이유는 분류기의 오류와 문서의 특성상 한 영역으로 규정짓기 어려운 문서가 존재하기 때문이다. 그러나 검색하고자 하는 사용자의 관심 분야나 찾고자 하는 성향이 다르더라도 시스템에 종속되지 않고 유연하게 대처하며 검색 성능(retrieval effectiveness)을 향상시킬 수 있다.사용되고 있어 적응에 문제점을 가지기도 하였다. 본 연구에서는 그 동안 계속되어 온 한글과 한잔의 사용에 관한 논쟁을 언어심리학적인 연구 방법을 통해 조사하였다. 즉, 글을 읽는 속도, 글의 의미를 얼마나 정확하게 이해했는지, 어느 것이 더 기억에 오래 남는지를 측정하여 어느 쪽의 입장이 옮은 지를 판단하는 것이다. 실험 결과는 문장을 읽는 시간에서는 한글 전용문인 경우에 월등히 빨랐다. 그러나. 내용에 대한 기억 검사에서는 국한 혼용 조건에서 더 우수하였다. 반면에, 이해력 검사에서는 천장 효과(Ceiling effect)로 두 조건간에 차이가 없었다. 따라서, 본 실험 결과에 따르면, 글의 읽기 속도가 중요한 문서에서는 한글 전용이 좋은 반면에 글의 내용 기억이 강조되는 경우에는 한자를 혼용하는 것이 더 효율적이다.이 높은 활성을 보였다. 7. 이상을 종합하여 볼 때 고구마 끝순에는 페놀화합물이 다량 함유되어 있어 높은 항산화 활성을 가지며, 아질산염소거능 및 ACE저해활성과 같은 생리적 효과도 높아 기능성 채소로 이용하기에 충분한 가치가 있다고 판단된다.등의 관련 질환의 예방, 치료용 의약품 개발과 기능성 식품에 효과적으로 이용될 수 있음을 시사한다.tall fescue 23%, Kentucky bluegrass 6%, perennial ryegrass 8%) 및 white clover 23%를 유지하였다. 이상의 결과를 종합할 때, 초종과 파종비율에 따른 혼파초지의 건물수량과 사료가치의 차이를 확인할 수 있었으며, 레드 클로버 + 혼파 초지가 건물수량과 사료가치를 높이는데 효과적이었다.\ell}$ 이었으며 , yeast extract 첨가(添加)하여 배양시(培養時)는 yeast extract

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.04b
• /
• pp.336-338
• /
• 2000

본 논문에서는 사용자의 적합 피드백을 기반으로 적합 문서들에서 발생하는 용어들과 초기 질의어간의 발생 빈도 유사도 및 퍼지 추론을 이용하여 용어의 가중치를 산정하는 방법에 대하여 제안한다. 피드백 문서들에서 발생하는 용어들 중에서 불용어를 제외한 모든 용어들을 질의로 확장될 수 있는 후보 용어들로 선택하고, 발생 빈도 유사성을 이용한 초기 질의어-후보 용어의 관련 정도, 용어의 IDF, DF 정보를 퍼지 추론에 적용하여 후보 용어의 초기 질의에 대한 최종적인 관련 정도를 산정 하였으며, 피드백 문서들에서의 가중치와 관련 정보를 결합하여 후보 용어들의 가중치를 산정 하였다.

• Journal of the Korea Society of Computer and Information
• /
• v.3 no.1
• /
• pp.35-46
• /
• 1998

In the Information Retrieval System. the relevance of retrieved items is determined by the judgement of the user and thus the evaluation of the system efficiency counts on the cognizance of users to some extent. The relevance feedback mechanism provides a device allowing iterative searches during which the query can be modified and refined based on user input from the relevant documents. The feedback system are generally reported to outperform non-feedback systems. The procedures and algorithms to implement the feedback mechanism are surveyed in this paper.

• Journal of the Korean Society for information Management
• /
• v.41 no.1
• /
• pp.261-282
• /
• 2024

This study proposed and evaluated electroencephalography (EEG)-based and eye-tracking-based methods to determine relevance by utilizing users' implicit relevance feedback while navigating content in a digital library. For this, EEG/eye-tracking experiments were conducted on 32 participants using video, image, and text data. To assess the usefulness of the proposed methods, deep learning-based artificial intelligence (AI) techniques were used as a competitive benchmark. The evaluation results showed that EEG component-based methods (av_P600 and f_P3b components) demonstrated high classification accuracy in selecting relevant videos and images (faces/emotions). In contrast, AI-based methods, specifically object recognition and natural language processing, showed high classification accuracy for selecting images (objects) and texts (newspaper articles). Finally, guidelines for implementing a digital library interface based on EEG, eye-tracking, and artificial intelligence technologies have been proposed. Specifically, a system model based on implicit relevance feedback has been presented. Moreover, to enhance classification accuracy, methods suitable for each media type have been suggested, including EEG-based, eye-tracking-based, and AI-based approaches.