• KIISE Transactions on Computing Practices
• /
• v.22 no.6
• /
• pp.290-295
• /
• 2016

Recently, folksonomy-based image-sharing sites where users cooperatively make and utilize tags of image annotation have been gaining popularity. Typically, these sites retrieve images for a user request using simple text-based matching and display retrieved images in the form of photo stream. However, these tags are personal and subjective and images are not categorized, which results in poor retrieval accuracy and low user satisfaction. In this paper, we propose a categorization scheme for folksonomy images which can improve the retrieval accuracy in the tag-based image retrieval systems. Consequently, images are classified by the semantic similarity using text-information and image-information generated on the folksonomy. To evaluate the performance of our proposed scheme, we collect folksonomy images and categorize them using text features and image features. And then, we compare its retrieval accuracy with that of existing systems.

• The Journal of the Korea Contents Association
• /
• v.7 no.7
• /
• pp.50-59
• /
• 2007

This paper describes a question analysis and expansion based on semantics for on efficient information retrieval. Results of all information retrieval systems include many non-relevant documents because the index cannot naturally reflect the contents of documents and because queries used in information retrieval systems cannot represent enough information in user's question. To solve this problem, we analyze user's question semantically, determine the answer type, and extract semantic features. And then we expand user's question using them and syntactic structures which are used to represent the answer. Our similarity is to rank documents which include expanded queries in high position. Especially, we found that an efficient document retrieval is possible by a question analysis and expansion based on semantics on natural language questions which are comparatively short but fully expressing the information demand of users.

• The Journal of Society for e-Business Studies
• /
• v.13 no.2
• /
• pp.213-223
• /
• 2008

The efficient discovery of services from a large scale collection of services has become an important issue[7, 24]. We studied a syntactic method for Web service discovery, rather than a semantic method. We regarded a service discovery as a retrieval problem on the proprietary XML formats, which were service descriptions in a registry DB. We modeled services and queries as probabilistic values and devised similarity-based retrieval techniques. The benefits of our way are follows. First, our system supports ranked service retrieval by keyword search. Second, we considers both of UDDI data and WSDL definitions of services amid query evaluation time. Last, our technique can be easily implemented on the off-theshelf DBMS and also utilize good features of DBMS maintenance.

• The Transactions of the Korea Information Processing Society
• /
• v.2 no.5
• /
• pp.711-720
• /
• 1995

This paper describes in object-oriented retrieval framework that is generally designed to store and retrieve the reusable components from the library regardless of the underlying representation of the library. We propose a retrieval framework on visual space so that reuser can identify their location at the library without any previous information of library structure. They can decide the directions of retrieval with the results displayed on the visual space and interact with the library using the defined simple retrieval operation that can assess the library information object. For doing this, 4I model was proposed. Librarian as well as reuser can easily construct the new library on the visual environment. It is the process to give the semantic of the information object. This paper discusses the basic concepts of our 4I model and explains each constituent of our model and shows a simple example of the system.

• 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.

• Proceedings of the IEEK Conference
• /
• 2001.06c
• /
• pp.47-50
• /
• 2001

This paper took advantage of SVD (Singular value Decomposition) techniques of LSI(Latent Semantic Indexing) to grasp easily terminology distribution. Existent LSI did to static database, propose that apply to dynamic database in this paper. But, if dynamic applies LSI to database, updating problem happens. Existent updating way is Recomputing method, Folding-in method, SVD-updating method. Proposed QR decomposition method to show performance improvement than existent three methods in this paper.

• Proceedings of the Korea Multimedia Society Conference
• /
• 1998.04a
• /
• pp.273-277
• /
• 1998

인터넷의 발전으로 많은 웹 기반 정보 검색 시스템의 구축은 사용자들에게 다양한 정보를 제공하고 있으나, 방대한 정보의 양을 사용자들은 효율적으로 이용하지 못하고 있다. 그러므로 본 논문에서는 웹 기반 정보 검색 시스템에서 의미에 따른 문서 검색 기법을 제공하기 위해 의미 링크에 대해 알아보고, 의미 링크를 이용한 웹 기반 정보 검색 시스템인 불교 문화재 답사시스템을 웹과 데이터베이스 연동 방식에 따라 분류하여 구현해 본다. 앞으로의 연구 방향은 일반적인 정보 검색 시스템에서 로봇을 통하여 데이터들의 의미를 파악함으로써 의미에 따른 검색 기법을 제공하는 것이다.

• STIMA Bulletin
• /
• s.6
• /
• pp.27-34
• /
• 2007

정보검색서비스는 '사용자가 얼마나 편리하게 검색할 수 있는가'와 '검색 결과에 얼마나 만족하는가'가 중요한데, 이는 정보검색 기술 개발에서 가장 중요하게 고려해야 할 사항이다. 본고는 과학기술 지식정보를 대상으로, 어휘망과 온톨로지적 성격을 가지고 있는 U-WIN을 기반으로 의미적인 정보검색 서비스가 가능하도록 하기 위하여, 한국과학기술정보연구원(KISTI)에서 개발 중인 U-WIN을 이용한 의미적 정보검색 기술과 시범서비스인 KISTI-STA를 소개한다.

• Proceedings of the CALSEC Conference
• /
• 2004.02a
• /
• pp.302-309
• /
• 2004

·Catalogs -information about products and services -Contents ＋ Classification Schema ＋ Operational Issues ·What do we do with them？ -[Schulten, et a;, 2001] ‥Narrow down search for complete set of applicable products ‥Comprehend individual description to the precision needed -Support other applications that use product information ‥SCM, ERP, e-Procurement, etc. ·Catalog Management System -Design, storage, navigation ＆ retrieval, transformation, communication, publication(omitted)

• Science of Emotion and Sensibility
• /
• v.21 no.3
• /
• pp.61-72
• /
• 2018

Items related to a crime that are known only to criminals and investigators can be used in the concealed information test (CIT) to assess whether the suspect is guilty of the offense. However, in many cases wherein the suspect is exposed to information about the crime, the CIT cannot be used. Although the perpetrator's memories about the details of the crime are episodic, the memories of a suspect who has inadvertently discovered the details of the crime are more likely to be semantic. The retrieval of episodic memories is associated with theta wave activity, whereas that of semantic memories is associated with alpha wave activity. Therefore, these aspects of memory retrieval can be useful in identifying the perpetrator of the crime. In this study, P300-based CITs were conducted in a guilty participant in a mock crime and an innocent participant who has been given information about the simulated offense. The results demonstrate that the difference in P300 amplitudes between the probe and the irrelevant stimulus did not differ between the guilty and innocent conditions. As expected, the lower theta band power (4-6 Hz) was higher in the probe than in the irrelevant stimulus in the guilty condition, but there was no difference in the innocent condition. Conversely, the upper alpha band power (8-10 Hz) was lower in the probe than in the irrelevant stimulus in the innocent condition, but there was no difference in the guilty condition. The possibility of using theta and alpha band powers in lie detection is discussed.