• Annual Conference on Human and Language Technology
• /
• 2004.10d
• /
• pp.283-289
• /
• 2004

조응어 해석(Anaphora Resolution)은 여러 응용 분야에 적절히 적용됨으로써 자연어 이해에서의 양적. 질적 향상을 가져 올 수 있다. 구어에서 조응어(anaphor)는 화자와 청자간에 공유된 정보를 간략하게 나타내는 형태이다. 특히, 구어를 주된 사용 언어로 하는 대화 시스템에서의 조응어의 양상을 살피고 해석하는 문제는 인간과 기계간의 의사소통(man-machine communication)을 온전히 하는 과제이다. 이 논문에서는 인간과 기계간의 대화 시스템(man-machine dialogue system)에서 흔히 쓰일 수 있는 조응 표현에 중점을 두어 그 유형을 살피고 해석의 방식을 제시하도록 한다.

• Koreanishche Zeitschrift fur Deutsche Sprachwissenschaft
• /
• v.5
• /
• pp.53-65
• /
• 2002

독일어의 관계대명사는, 선행사의 반복과 그에 결부된 일치 규정을 철저하게 따름으로써 가능한 한, 언어의 모호성을 배제하려 한다. 이에 반해 한국어에서는 그런 것이 전혀 필요하지 않다. 독일어의 관계대명사에 해당하는 부가어 문장이 항상 그의 관계어/선행사 앞에 위치하기 때문이다. 이 때문에 양측의 (모국어)화자는 심각한 언어간섭 현상에 빠지기 마련이다 그들의 차이가 서로에게 너무 큰 까닭이다. 이러한 대립적인 어려운 문제들은, 일면에서는 전치성과 후치성이란 언어 특성을 익히고, 다른 일면에서는 한국인이 번역에서 독일어의 관계대명사에 결부된 구조를 완전히 생략함으로써 극복될 수 있다. 독일어나 한국어처럼 상호 대립적으로 발달한 언어를 정복하기 위해서, 가장 좋은 방법으로는 문장 자체를 한 단어처럼 익히는 것이다. 이것은 또한 서술을 위한 표현능력을 신장하는데도 효과적이다. 이에 따른 문법을 따로 배워야 한다는 생각은 전혀 필요하지 않다. 왜냐하면 언어습득장치는 모든 것을 언제나 문법적/체계적/분석적으로 수용하기 때문이다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.3 no.1
• /
• pp.197-208
• /
• 1999

In this paper, we proposed a multimedia hardware architecture for video-conferencing in view of the multimedia data flow. By simulating the architecture model, we analyzed the bottleneck of multimedia data flow, varying video size, frame rate, number of participants, and video data compression rate. To confirm the simulation results, we also implemented and tested the architecture that almost includes the analyzed requirements for video- conferencing. From the analysis of I/O system, we found the considerations in designing a multimedia I/O system.

• The Transactions of the Korea Information Processing Society
• /
• v.3 no.3
• /
• pp.608-616
• /
• 1996

In this paper, we propose a multimedia hardware architecture for desktop video-conferencing in view of the multimedia data flow and analyze system operation using queueing model. We analyze the bottleneck of multimedia data flow for video- conferencing by simulation, varying the video size, frame rate, the number of participants, and video data compression rate, And we also implement and test the archiecture, that almost includes the analyzed requirements for video-conferencing, to confirm the simulation results. This paper describes the considerations in designing a multimedia I/O system.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2019.01a
• /
• pp.391-393
• /
• 2019

본 논문은 일상적인 대화로부터 감성을 추출하고 분석함으로써 상황에 맞는 대화의 내용과 분위기를 이미지의 색상으로 표현할 수 있는 이미지 색상 변환 프레임워크를 소개한다. 본 연구는 라즈베리 파이와 마이크 센서를 기반으로 사용자로부터 목소리를 입력받을 수 있는 모듈을 제작하고, 그 목소리로부터 감성을 분석한다. 분석된 감성을 이용하여 이미지의 색상을 자동으로 변환하는 기술과 통합함으로써 청각장애인 및 미취학 아동들이 화자의 대화를 이미지를 통해 쉽게 인지하여 의사소통 및 감성 전달 환경을 개선하고자 한다.

• Journal of KIISE
• /
• v.44 no.12
• /
• pp.1252-1257
• /
• 2017

In dialogues, speakers' intentions can be represented by sets of an emotion, a speech act, and a predicator. Therefore, dialogue systems should capture and process these implied characteristics of utterances. Many previous studies have considered such determination as independent classification problems, but others have showed them to be associated with each other. In this paper, we propose an integrated model that simultaneously determines emotions, speech acts, and predicators using a convolution neural network. The proposed model consists of a particular abstraction layer, mutually independent informations of these characteristics are abstracted. In the shared abstraction layer, combinations of the independent information is abstracted. During training, errors of emotions, errors of speech acts, and errors of predicators are partially back-propagated through the layers. In the experiments, the proposed integrated model showed better performances (2%p in emotion determination, 11%p in speech act determination, and 3%p in predicator determination) than independent determination models.

• Korean Journal of Logic
• /
• v.16 no.1
• /
• pp.61-85
• /
• 2013

The author, after introducing the concept of de re belief and discussing de re/de dicto ambiguity in belief context and modal context, concludes that modal sentences of Korean language does not show any distinctive traits against English. He, after discussing this ambiguity in negative sentence a la Russell, tries to show that Korean provides two way of negation construction, one of which corresponds to de re negation (primary occurrence in Russell's terms). De re reading makes referentially transparent context, thus permits substitutions of identicals salva veritate; De dicto reading does not. Korean ending 'rago', used with quotation verbs, speech act verbs and cognitive attitude verbs, deserves some attention in that it permits de re sentences in addition to de re/de dicto ambiguous sentences. 'Rago' also makes speaker's commitment to the content of the intensionally contained clause 'neutral', in contrast with other Korean endings such as 'um/im' and 'raneun gut' which make speaker's positive commitment. This explains why the maxim of western epistemology that knowledge presupposes truth does not hold in Korean 'rago' sentences.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.23 no.2
• /
• pp.184-189
• /
• 2013

In this paper, we propose the injurious moving picture filtering system using certain sounds contained in the injurious moving picture to filter injurious moving picture which is distributed without limitation in internet and internet storage space. For this purpose, the Gaussian Mixture Model which can well represent the characteristics of the sound, is used and frame level likelihood estimation is used to calculate the likelihood between filtering target data and the sound models. Also, the pruning method which can real-time proceed by reducing the comparing number of data, is applied for real-time processing, and MWMR method which showed good performance from existing speaker identification, is applied for the distinguish performance of high precision. In the identification experiment result, in case of the frame rate which is the proportion of total frame to high likelihood frame, is set to 50%, identification error rate is 6.06%, and in case of frame rate is set to 60%, error rate is 3.03%. As the result, the proposed system can distinguish between general and injurious moving picture effectively.

• 한국HCI학회:학술대회논문집
• /
• 2009.02a
• /
• pp.129-134
• /
• 2009

Recently, percentage of people who use SMS service is increasing. However, it is difficult to express own complicated emotion with text and emoticon of exited SMS service. This paper focuses on that point and practical uses character animation to express emotion and nuance correctly, funny. Also this paper suggests emotion based gesture animation generation system that use character's facial expression and gesture to delivery emotion excitably and clearly than only speaking. Michel[1] investigated interview movies of a person whose gesturing style they wish to animate and suggested gesture generation graph for stylized gesture animation. In this paper, we make focus to analyze and abstracted emotional gestures of Disney animation characters and did 3D modeling of these emotional gestures expanding Michel[1]'s research. To express emotion of person, suggests a emotion gesture generation graph that reflects emotion flow graph express emotion flow for probability. We investigated user reaction for research the propriety of suggested system and alternation propriety.

• Journal of the HCI Society of Korea
• /
• v.12 no.1
• /
• pp.5-13
• /
• 2017

Interactive intervention of the human presenter is one of the important factors that make the visualization more effective. Rather than just showing the content, the presenter enhances the process of the information delivery by providing the context of visualization. In this paper, we define this as an augmented presentation. In augmented presentation concept, the presenter can facilitate presentation more actively by being fully immersed in the visualization space and reaching and interacting into digital information. In order to concrete the concept, we design presentation space that enables the presenter to be seamlessly immersed in the visualization. Also we increase the presenter's roles as a storyteller, controller and augmenter allowing the presenter to fully support communicative process between the audience and the visualization. Then, we present an augmented presentation system to verify the proposed concept. We rendered 3D visualization through a half-mirror film and a wall projection screen that are place in parallel and applied with stereoscopic images, then, spatially align the presenter inside the virtual visualization space. After that, we conduct a controlled experiment to investigate the subjective level of immersion and engagement of the audience to HoloStation compared to traditional presentation system. Our initial investigation suggests that the newly conceived augmented presentation has potential not only to enhance the information presentation but also to supports the delivery of visualization.