• Journal of Institute of Control, Robotics and Systems
• /
• v.14 no.5
• /
• pp.473-482
• /
• 2008

In this paper we propose an expert system for content-based image retrieval with object database. The proposed system finds keyword by using knowledge-base and feature of extracted object, and retrieves image by using keyword based image retrieval method. The system can decrease error of image retrieval and save running time. The system also checks whether similar objects exist or not. If not, user can store information of object in object database. Proposed system is flexible and extensible, enabling experts to incrementally add more knowledge and information. Experimental results show that the proposed system is more effective than existing content-based image retrieval method in running time and precision.

• Journal of the Korea Society of Computer and Information
• /
• v.11 no.6 s.44
• /
• pp.27-33
• /
• 2006

In this thesis, we presents the component design and retrieval prototype system which can retrieve the most appropriate component for reuse system. If it used only an abject to reuse unit, it can not understand mutual reaction and can not sustain relationship between objects. Therefore, we porposed clustering component for reuse unit. It makes to understand relationship and reaction between object, so User can retrieve the proper component for reuse system. And porposed retrieval prototype system can select the correct things. it provide 3 facet retrieval environment against ambiguous retrieval definition.

• The KIPS Transactions:PartD
• /
• v.8D no.5
• /
• pp.483-494
• /
• 2001

Most of hypertext retrieval models consider documents as independent entities. They ignore relationships between documents of link semantics. in an information retrieval system for hypertext documents, retrieval effectiveness can be improved when ling information is used. Previous link-based hypertext retrieval models ignore link information while indexing. They utilize link information to re-rank the retrieval results. Therefore they are limited that only the documents is result-set utilize link information. This paper utilizes link information when indexing. We present how to use term weighting and inLinks weighting for ranking the relevant documents. Experimental results show that recall and precision evaluation according to the link semantics and the comparison with previously link_based hypertext retrieval model.

• Journal of Korean Library and Information Science Society
• /
• v.31 no.2
• /
• pp.279-302
• /
• 2000

This study is to develop a new automatic system for the Korean thesaurus browser by which we can automatically control all the processes of searching queries such as, representation, generation, extension and construction of searching strategy and feedback searching. The system in this study is programmed by Delphi 4.0(PASCAL) and consists of database system, automatic indexing, clustering technique, establishing and expressing thesaurus, and automatic information retrieval technique. The results proved by this system are as follows: 1)By using the new automatic thesaurus browser developed by the new algorithm, we can perform information retrieval, automatic indexing, clustering technique, establishing and expressing thesaurus, information retrieval technique, and retrieval feedback. Thus it turns out that even the beginner user can easily access special terms about the field of a specific subject. 2) The thesaurus browser in this paper has such merits as the easiness of establishing, the convenience of using, and the good results of information retrieval in terms of the rate of speed, degree, and regeneration. Thus, it t m out very pragmatic.

• Journal of Korea Society of Industrial Information Systems
• /
• v.8 no.4
• /
• pp.31-37
• /
• 2003

This paper is a review of the large-scale information retrieval system, CONDOR. This system was developed by the consortium that consists of Chonbuk National University, Searchline Co. and Carnegie Mellon University. This system is based on the probabilistic model of information retrieval systems. The multi-language query processing, online document summarization based on query and dynamic hierarchy clustering of this system make difference of other systems. We test this system with 30 million web documents successfully.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2006.11a
• /
• pp.227-230
• /
• 2006

A content-based image retrieval system using MPEG-7 is designed and implemented in this thesis. The implemented system uses existing MPEG-7 Visual Descriptors. In addition, a new descriptor for efficient retrieval of bio images is proposed and utilized in the developed content-based image retrieval system. Comparing proposed CBSD(Compact Binary Shape Descriptor) with Edge Histogram Descriptor(EHD) and Region Shape Descriptor(RSD), it shows good retrieval performance in NMRR. The proposed descriptor is robust to large modification of brightness and contrast and especially improved retrieval performance to search images with similar shapes. Also proposed system adopts distributed architecture to solve increased server overload and network delay. Updating module of client efficiently reduces downloading time for metadata. The developed system can efficiently retrieve images without causing server's overload.

• Journal of the Korea Society for Simulation
• /
• v.19 no.1
• /
• pp.83-90
• /
• 2010

We describe a novel method for improving the performance of the UCC retrieval using content reputation and user reputation. The UCC retrieval is a part of the information retrieval. The goal of the information retrieval system finds documents what users want, so the goal of the UCC retrieval system tries to find UCCs themselves instead of documents. Unlike the document, the UCC has not enough textual information. Therefore, we try to use the content reputation and the user reputation based on non-textual information to gain improved retrieval performance. We evaluate content reputation using the information of the UCC itself and social activities between users related with UCCs. We evaluate user reputation using individual social activities between users or users and UCCs. We build a network with users and UCCs from social activities, and then we can get the user reputation from the network by graph algorithms. We collect the information of users and UCCs from YouTube and implement two systems using content reputation and user reputation. And then we compare two systems. From the experiment results, we can see that the system using content reputation outperforms than the system using user reputation. This result is expected to use the UCC retrieval in the feature.

• Journal of the Korean Society for Library and Information Science
• /
• v.45 no.2
• /
• pp.345-366
• /
• 2011

The law affects our daily lives, and hence, constitutes a crucial information resource. However, electronic access to legal information using keyword-based retrieval systems appears to provide users with limited satisfaction. There are many factors behind this inadequacy. First, the discrepancies between formal legal terms and their counterparts in common language are quite large. Second, the situation is further confounded by frequent abbreviations in legal terms. Third, even though there is a constant deluge of legal information, users' needs have evolved to demand more Q and A type searches. All of these factors make the existing retrieval systems inefficient and ineffective. This article suggests an ontology-based system as a means to deal with such difficulties. To that end, a legal retrieval system(experimental system), built on the basis of a newly-constructed law ontology, was tested against a keyword-based legal retrieval system(existing one), yielding data on their relative effectiveness in retrieval and user satisfaction.

• Journal of KIISE:Software and Applications
• /
• v.29 no.1_2
• /
• pp.30-41
• /
• 2002

We focus on the mobile agents which are considered as new paradigm to solve information retrieval of large volumes of data in the distributed and heterogeneous environment. The mobile agent moves the computation to data instead of large volumes of data to computations. In this paper, we propose an information retrieval model, which can effectively search data in the distributed and heterogeneous environment, using mobile agents. Our model is applied to the design and implementation of an Q&A(Question and Answer) retrieval system. Our Q&A retrieval system, called QASSMA(Q&A Search System using Mobile Agents), uses mobile agents to retrieve articles from Q&A boards and newsgroups that exist in the heterogeneous and distributed environment. QASSMA has the following features and advantages. First, the mobile retrieval agent moves to the destination server to retrieve articles to reduce the retrieval time by eliminating data traffics from the server to the client host. Also it can reduce the traffic that was occurred in the centralized network system, and reduce the usage of resources by sending its agent and running in the destination host. Finally, the mobile retrieval agent of QASSMA can add and update dynamically the class file according to its retrieval environment, and support other retrieval manner. In this paper, we have shown that our Q&A retrieval system using mobile agents is more efficient than the retrieval system using static agents by our experiments.

• Journal of the Korean Society for information Management
• /
• v.2 no.1
• /
• pp.108-135
• /
• 1985

When an information retrieval system is implemented on-line, users can interact with the system to improve the searches. There are studies which achieved dramatic improvements in system effectiveness by using automatic relevance feedback, a technique for reformulating a patron query based on initial retrieval result. In this thesis, an automatic query modification model was applied to a controlled keyword system.