• The Transactions of the Korea Information Processing Society
• /
• v.4 no.4
• /
• pp.963-972
• /
• 1997

In imformation retrival, query reformulation algorithms construct querise from a set of intial input and feedback documents, and retrieval performance cna be varied by different sets of input documents. In this study, we developed a criterion for measuring the performance sensitivity of query reformulation algorithms to unput sets. In addition, we also propose a way of mesuring the changes in retrived area, (CIRA) during qucry reformulation. We cimpared CIRAs of query refromulation algorithms (i.e., query tree, DNF method, and Dillon's method) using three test sets:the CACM, CISI, and Medlars. In the experiments, the query tree showed the highest decreasing CIRA during refirmulations, which means the fastest convergence rate to an output set. For sensitivity analysis, the query tree sored the highest sensitivity to different input sets even though its differences to the tther algorithms are very small.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.3
• /
• pp.709-719
• /
• 1997

Performance of an algorithm can be mesaurde from serval aspects.Suppose thre is a query formulation al-gorithm.Even though this algorithm shows high retrival performance, ie, high recall and percision, retriveing items can rake a long time.In this study, we time complexity of automatic query reformulation algorithms, named the query Tree, DNF method, and Dillon's method, and comparethem in theoretical and practical aspects using a tral-time performance)the absolute times for each algorithm to fromulate a query)in a Sun SparcStation 2. In experiments using three test sets, CSCM, CISI, and Medlars, the query Tree algorithm was the fastest among the three algorithms tested.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.5
• /
• pp.1297-1307
• /
• 1997

This paper presents a multimedia mail system which transmit and redeive multimedia mailing messges on Intemet.This mail system is an extension of the exsting e-mail system for multimedia uncluding,text,image,MPEG video,and binary data,The MIME(Multipurpose Intert Mail Extensions)format,which is an extension of REF-822 maill format,is used to reprssent multimedia,and SMTP(Simple Mail Transfer Protocol)is utilized as a mail transport prttocol.The IMAP(Intenet Mail Access Protcol)which privides more functions than the widely used POP(Post Office Protocol)is used as a mailbox retrival protocol.The mail client is complemented on a multimedia PC while the server is implemented on a UNIX system.In the mail system, a mail sending program allows a user to attach binary files such as Postscript files and MPEG compressed video,while a receiving program provides direct interface to application programs to play back received multimedia mail messages.

• Journal of KIISE:Computing Practices and Letters
• /
• v.6 no.5
• /
• pp.496-504
• /
• 2000

The Web gateway is key technology for inter-operating WWW and databases. The previous KRISTAL-II information retrieval system, developed by KORDIC(Korea Research & Development Information Center), used a simple CGI structure web gateway. While the simple CGI structure web gateway is easy to implement, it is not suitable for processing a large number of on-line retrival requests. When considering the growth of the Internet and WWW, it is very important to develop a web gateway efficiently supporting a large number of concurrent users. In this paper, we propose a 3-tier client-server structure web gateway for the KRISTAL-II information system. We also evaluate the performance of the proposed web gateway through experiments.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.4
• /
• pp.895-907
• /
• 1997

The mining of assciation rules disovers the trndency of events ocuring simultaneously in large databases. Previous announced research on association rules deals with associations with associations with respect to the whole transaction. However, xome association rules could have very high confidence in a sub-range of the time domain, even though they do not have quite high confidence in the whole time domain. Such kind of association rules are ecpected to be very usdful in various decion making problems.In this paper, we define transient association rule, as an association with high cimfidence worthy of special attention in a partial time interval, and propose an dfficeint algorithm wich finds out the time intervals appropriate to transient association rules from large-databases.We propose the data-driven retrival method excluding unecessary interval search, and design an effective data structure manageable in main memory obtined by one scanning of database, which offers the necessary information to next retrieval phase. In addition, our simulation shows that the suggested algorithm has reliable performance at the time cost acceptable in application areas.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.4
• /
• pp.933-943
• /
• 1997

In this paper, a conernt-based image retrival system with color assigments has been stueide and implment-ed. The color of images has been extracted after changing RGB color space to HSV(hue, saturation, value)that is the most compatible color for peop]e's feeling. In the color extracting, an image is divided into 9 different areasand 3 major colors for each area are selected by using color histograms. It is possible to chose the class of umages by keywords. We are evaluate four different types of queries such as an image input, keywords with color assignments, combining an image input and keywords with color assinments, and selecting specific part of an umage. Experimental rusults show that four different query types privide precision/recall 0.55/0.37, 0.57/0.43, 0.59/0.45 and 0.63/0.61, respectively. With color assignments, the retrieval system has been able to obtain high performance and validity.

• Korean Journal of Oriental Medicine
• /
• v.18 no.3
• /
• pp.133-139
• /
• 2012

Objective : UMLS is 'Ontology' which establishes the database for medical terminology by gathering various medical vocabularies representing same fundamental concepts. Method : Although Chinese character are represented in the Chinese part of Korean Unicode system in a computer, writing of Chinese characters is vary depending on Chinese input systems and Chinese writers' levels of knowledge. As the result of this, representation of Chinese writing in a computer will be considerably different from an old Chinese document. Therefore, a meaningful relationship between digital Chinese terminology and translated Korean is necessary in order to build Ontology for Chinese medical terms from Oriental medical prescription in a computer system. Result : This research will present 1:1 mapping information among the Chinese characters used in the Oriental medical prescription with analysis of 'same character different sound' and 'same meaning different shape' in Chinese part of Unicode systems. Conclusions : Furthermore, the research will provide top-down menu of relationship between Chinese term and Korean term in medical prescription with assumption of that the Oriental medical prescription has its own unique meaning.

• Journal of the Korea Convergence Society
• /
• v.13 no.1
• /
• pp.13-21
• /
• 2022

Recent deep learning-based research shows excellent performance in most natural language processing (NLP) fields with pre-trained language models. In particular, the auto-encoder-based language model proves its excellent performance and usefulness in various fields of Korean language understanding. However, the decoder-based Korean generative model even suffers from generating simple sentences. Also, there is few detailed research and data for the field of conversation where generative models are most commonly utilized. Therefore, this paper constructs multi-turn dialogue data for a Korean generative model. In addition, we compare and analyze the performance by improving the dialogue ability of the generative model through transfer learning. In addition, we propose a method of supplementing the insufficient dialogue generation ability of the model by extracting recommended response candidates from external knowledge information through a retrival model.

• The KIPS Transactions:PartD
• /
• v.9D no.6
• /
• pp.1055-1062
• /
• 2002

Recently, the lesson plan document that is imported in the educational area is not provided to the educational information systematically, and the teachers are not easy to compose the lessen plan documentation. So, it needs additional time and effort to develope the lesson plan documents. Because of increasing the distributing network. web-based lesson plan system is required to all of the education area. Therefore, we need to compose the lesson plan that is possible to obtain the various teacher's requirement by providing creation, retrival, and reusability of document using the standard XML on web. In this paper, we developed the system for creating the common DTD (Document Type Definition), providing the standard XML document through the common DTD over the lesson plan analysis. In this system, it provides the editor to compose the lesson plan and supports the searching function to improvement of reusability on the existing lesson plan. We design the searching functions such as the structure base, facet and keyword. The composed lesson plans are interoperated with Database. Consequently, we can share the information on web by composing the lesson plan using the XML and save the time and cost by directly writing the lesson plan on web. We can also provide the improved learning environment.

• Korean Journal of Cognitive Science
• /
• v.24 no.2
• /
• pp.111-132
• /
• 2013

Previous research has shown that aging asymmetrically affects various functions in language. It is known that older adults show deficits in language production compared to young adults, while the performance in semantic processing is similar between older and young adults. The tip-of-the-tongue (TOT) phenomenon effectively reflects failure in retrieval of phonological knowledge. Older adults report TOTs more often than young adults and the cause of this phenomenon has been explained by two frameworks: the 'blocking hypothesis' and 'transmission deficit hypothesis'. This study examines the effect of aging on the retrival of phonological knowledge by inducing TOTs in the laboratory. Two variables were manipulated: age and word category. Participants were young and older adults, and stimuli was selected from 5 categories of words. After the participants read a definition about a target word, they reported three conditions: 'know', 'don't know', 'TOT'. The results were as follows: First, the older adults reported TOTs more often than the young adults. Second, TOTs occurred more in proper nouns such as names of persons and places. Third, in the category that TOTs occurred more often, there was a bigger age difference. Fourth, older adults reported fewer alternative words during TOT than young adults. Fifth, participants tended to report the partial information during TOT in characters. These results show the age-related difficulty in the retrieval of phonological knowledge in Korean. It is explained by the transmission deficit hypothesis and the characteristics of Korean orthography and phonology.