• Annual Conference on Human and Language Technology
• /
• 2002.10e
• /
• pp.9-15
• /
• 2002

이 연구는 형태 분석 정보가 부착된 말뭉치를 구축할 때 합성어를 처리하기 위한 방법론을 제시하고, 그 타당성을 검증해 보는 데 있다. 그동안 합성어 처리를 위해서 합성어 선정 기준을 이용하거나 목록을 이용하는 방법이 이용되었는데, 본고에서는 ${\ulcorner}$표준국어대사전${\lrcorner}$의 합성어 목록을 참조하는 것이 적절한 방법이 될 수 있음을 보이고자 한다. 또한 이 방법을 실제 말뭉치 구축에 활용할 경우, 원문의 띄어쓰기 정보가 합성어 처리에서 중요한 요인이 될 수 있다는 점을 지적하고, 이러한 처리가 가지는 한계와 의의에 대해서도 논의하고자 한다.

• Proceedings of the Korean Society of Fisheries Technology Conference
• /
• 2003.05a
• /
• pp.191-192
• /
• 2003

수산양식에 있어서 초기 치자어의 영양은 치자어의 생존 및 성장에 많은 영향을 미친다. 이와 같은 영양에는 주로 DHA와 EPA 등으로 많은 연구자들은 그 초점을 두고 있다. 그러나, 실제 어류에 있어서 부화 직후의 영양소 섭식 능력은 많은 종에 있어서 미지수로 종묘생산시 초기 감모기를 겪고 있으며 이 시기를 부화 후 first feeding 단계의 5∼10일로 잡고 있다. 그러나 이 초기 감모기를 지난 자치어는 rotifer와 Artemia 같은 살아있는 먹이생물을 이용하기에 충분하기 때문에 생존율에는 큰 영향이 없는 것으로 보고되어지고 있다. (중략)

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.12
• /
• pp.1254-1258
• /
• 2010

In information retrieval, different surface forms of the same object can cause poor performance of systems. In this paper, we propose the method extracting alternative words using translation words as features of each word extracted from parallel corpus, korean/english title pair of patent information. Also, we propose an association word filtering method to remove association words from an alternative word list. Evaluation results show that the proposed method outperforms other alternative word extraction methods.

• Journal of KIISE:Databases
• /
• v.30 no.2
• /
• pp.109-124
• /
• 2003

Although most commercial relational DBMSs Provide a graphical query language for the user friendly interfaces of the databases, few research has been done for graphical query languages in object databases. Expressing complex query conditions in a concise and intuitive way has been an important issue in the design of graphical query languages. Since the object data model and object query languages are more complex than those of the relational ones, the graphical object query language should have a concise and intuitive representation method. We propose a graphical object query language called GOQL (Graphical Object Query Language) for object databases. By employing simple graphical notations, advanced features of the object queries such as path expressions including set attributes, quantifiers, and/or methods can be represented in a simple graphical notation. GOQL has an excellent expressive power compared with previous graphical object query languages. We show that path expressions in XSQL(1,2) can be represented by the simple graphical notations in GOQL. We also propose an algorithm that translates a graphical query in GOQL into the textual object query with the same semantics. We finally describe implementation results of GOQL in the Internet environments.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.4
• /
• pp.943-950
• /
• 2011

In this paper, we presented an attention measurement method in 2D and 3D image/video to be applied for image and video retargeting and compressive processing. 2D attention is derived from the three main components, intensity, color, and orientation, while depth information is added for 3D attention. A rarity-based attention method is presented to obtain more interested region or objects. Displaced depth information is matched to attention probability in distorted stereo images and finally a stereo distortion predictor is designed by integrating low-level HVS responses. As results, more efficient attention scheme is developed from the conventional methods and performance is proved by applying for video retargeting.

• Annual Conference on Human and Language Technology
• /
• 2001.10d
• /
• pp.294-301
• /
• 2001

기계번역이 인간의 언어 능력을 기계로 구현한다는 점에서 전산학적 성격이 강하다면, 변환 사전은 인간의 어휘부(lexicon) 정보를 그대로 기계에 표상한다는 점에서 언어학적 성격이 강하다. 여기서는 다양한 어휘부 정보 중에서 한영 기계번역에서 필요한 언어학적 정보를 추출하고 이러한 정보를 바탕으로 적절한 동사 대역어 선택을 위한 변환 사전의 모형을 만들어 보고자 하였다. 한영 기계번역에서 적절한 동사 대역어 선택의 어려움은 한국어 동형어 처리 문제와 한국어에서는 포착되지 않지만 영어로 번역하는 과정에서 발생하는 영어 표현의 특수성 때문에 기인한 것으로 볼 수 있다. 이 논문에서는 이러한 문제를 논항과 문법 형태소, 선택제약, 개별 어휘 등의 기초적인 언어학적 개념을 이용한 변환사전을 통해 해결한다. 또한 동사 대역어 선택에 영향을 미치는 이러한 개별적인 요인들은 실제 변환사전의 기술에 있어서는 복합적으로 적용됨을 동사 '먹다'의 기술을 통해 확인할 수 있다.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.10
• /
• pp.1005-1009
• /
• 2010

Elliptical and anaphoric utterances occur frequently during spoken dialogue. Because discourse analysis rests on the basic premise that linguistic items cannot be understood without reference to the context, ellipsis and anaphora resolution plays an important role in discourse analysis. In this paper, we present a spoken dialogue system improving the robustness at dialogue level based on discourse analysis, such as anaphora and ellipsis resolution. The applicability and effectiveness of the proposed method is evaluated in the TV domain.

• Korean Journal of Cognitive Science
• /
• v.14 no.2
• /
• pp.49-56
• /
• 2003

In this article, we investigate some structural properties of head-internal relative clauses in Korean and their development in Korean-speaking children. In this study, we found head-internal relative clauses in Korean has a more limited domain than head-external relative clauses with respect to positions that can be relativized and clause boundaries that the head can move across: in head-internal relative clauses, only a subject or an object can be relativized and doubly embedded clause constitutes a barrier to the movement of the head. We also found that head-internal relative clauses are easier to understand and produce than head-external relative clauses. In addition, we found that head-internal relative clauses emerge earlier than head-external relative clauses in the acquisition of relative clauses in Korean. The preference of young children (aged 4 and 5) for head-internal relative clauses over head-external relative clauses suggests that children like to use head-internal relative clauses at an early developmental stage when they have difficulty in using head-external relative clauses.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.21 no.4
• /
• pp.414-422
• /
• 2011

IR (Information Retrieval) systems have the methods that compare relationships between query and index to identify document that may be fit to the user's query keyword. However, the methods usually ignore the importance of relations that are not expressed in the query. Therefore, in this study, we describe how to refine the queries' relation from keyword and to reveal the hidden intent. A useful relationship between query and keyword in IR wth studied and we classified the tion fromrelation. Firstfromall, we did researchmrelated on semantic relationship and ontolhiical researchmin foreign and domestic research, and also analyzed semantic network practices, information retrieval technolhiy, extracted and classified the tion fromrelationships s' relasite's real-world datamin whichminformation retrieval technolhiin fare applied. Next, we souiht to solve the problems occurred frequently i' relasituation that searchers tioically face. I' relacurrent search technolhiy, the mesh searchmresult fare poured by simply comparn ina query with index terms. Therefore, the need for an intelligent search fittn inusers' intent is required. The relationships between two queries to re hiddee and identify relasearcher's intent have to be revealed. By analyzn inthe practical cthes s' queries and classifyn inthem into nine kind fromrelationship tion, we proposed the method to design relation revealn inand role namn i, and we have also illustrated limitations of that methods.

• KIISE Transactions on Computing Practices
• /
• v.20 no.9
• /
• pp.492-497
• /
• 2014

In this paper, we propose a method of retrieving opinioned tweets in Twitter, which is the one of the popular Social Network Services and shares diverse opinions among various users. In typical opinion retrieval systems, they may consider the presence of sentiment phrases (subjectivity) as the important factor even if the subjective phrases are not related to a given query or speaker. To alleviate these problems, we utilized the syntactic structure of a sentence to identify the relationships between 1) subjectivity-query and 2) subjectivity-speaker and 3) the syntactic role of subjectivity. Besides, our learning-to-rank approach is trained to retrieve opinioned tweets based on query-relevance, textual features, user information, and Twitter-specific features. Experimental results on real world data show that our proposed method can achieve better performance than several baseline methods in terms of precision and nDCG.