• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권8호
• /
• pp.4270-4284
• /
• 2019

This study presents a reference model (RM) and the architecture of a cognitive health advisor (CHA) that integrates information with ambient intelligence. By controlling the information using the CHA platform, the reference model can provide various ambient intelligent solutions to a user. Herein, a novel approach to a CHA RM based on evolutional cyber-physical systems is proposed. The objective of the CHA RM is to improve personal health by managing data integration from many devices as well as conduct a new feedback cycle, which includes training and consulting to improve quality of life. The RM can provide an overview of the basis for implementing concrete software architectures. The proposed RM provides a standardized clarification for developers and service designers in the design and implementation process. The CHA RM provides a new approach to developing a digital healthcare model that includes integrated systems, subsystems, and components. New features for chatbots and feedback functions set the position of the conversational interface system to improve human health by integrating information, analytics, and decisions and feedback as an advisor on the CHA platform.

• ETRI Journal
• /
• 제41권1호
• /
• pp.109-116
• /
• 2019

This paper explores new techniques that are based on a hidden-layer linear transformation for fast speaker adaptation used in deep neural networks (DNNs). Conventional methods using affine transformations are ineffective because they require a relatively large number of parameters to perform. Meanwhile, methods that employ singular-value decomposition (SVD) are utilized because they are effective at reducing adaptive parameters. However, a matrix decomposition is computationally expensive when using online services. We propose the use of an extended diagonal linear transformation method to minimize adaptation parameters without SVD to increase the performance level for tasks that require smaller degrees of adaptation. In Korean large vocabulary continuous speech recognition (LVCSR) tasks, the proposed method shows significant improvements with error-reduction rates of 8.4% and 17.1% in five and 50 conversational sentence adaptations, respectively. Compared with the adaptation methods using SVD, there is an increased recognition performance with fewer parameters.

• 수산해양기술연구
• /
• 제57권1호
• /
• pp.69-77
• /
• 2021

The noise environment was evaluated using the ISO recommended NR evaluation curve and PSIL (Preferred Sound Interference Level) in order to investigate the onboard educational environment according to the noise in the Motor Vessel Saehaerim, a fishing training ship under making way. As a result, NRNs were measured at 37-61 dB in accommodation areas, 44-56 dB in work areas, 37-57 dB in educational and conference areas, 83-103 dB in engine areas and 65.3 dB and 51.2 dB in the work and education areas respectively based on PSIL. The NRNs, which evaluated the cabin of the experimental ship according to the purpose, exceeded all of the indoor standard noise recommended by the ISO, and the PSIL had a generally short conversational distance within 0.25-2.3 m and 0.75-1.3 m for teaching and work areas.

• International Journal of Internet, Broadcasting and Communication
• /
• 제13권2호
• /
• pp.166-172
• /
• 2021

The purpose of this study is to identify the impact of informatization minds of college students majoring in sports on information development efforts and academic achievement. Accordingly, a survey was conducted on 200 college students majoring in physical education at D University, and 197 students were selected as the final effective samples. Data processing was conducted using SPSS 23 for frequency analysis, reliability analysis, correlation analysis, and regression analysis. The results are as follows: First, recognition of the necessity of informatization education, a subfactor of the informatization mind, has a positive impact on informatino development efforts. Second, recognition of the necessity of informatization education, understanding and utilization of informatization concepts among sub-factors of informatization minds have a positive impact on the major learning achievement. Third, all the conversational factors in the informatization mind have a positive impact on study problem resolution. Fourth, information development efforts have a positive impact on the major learning achievement and the study problem resolution. Therefore, efforts are needed to increase the information mindset of college students major in sports.

• 스마트미디어저널
• /
• 제11권6호
• /
• pp.51-63
• /
• 2022

Metaverse provides a simulated environment where a large number of users can participate in various activities. In order for Metaverse to be sustainable, it is necessary to study ethics that can be applied to a Metaverse service platform. In this paper, Metaverse ethics and the rules for applying to the platform are explored. And, in order to judge the ethicality of avatar actions in social Metaverse, the identity, interaction, and relationship of an avatar are investigated. Then, an action-based audit approach to avatar interactions (e.g., dialogues, gestures, facial expressions) is introduced in two cases that an avatar enters a digital world and that an avatar requests the auditing to subjects, e.g., avatars controlled by human users, artificial intelligence (AI) avatars (e.g., as conversational bots), and virtual objects. Pseudocodes for performing the two cases in a system are presented and they are examined based on the description of the avatars' actions.

• 전자통신동향분석
• /
• 제38권5호
• /
• pp.1-11
• /
• 2023

The emergence of generative hyperscale artificial intelligence (AI) has enabled new services, such as image-generating AI and conversational AI based on large language models. Such services likely lead to the influx of numerous users, who cannot be handled using conventional AI models. Furthermore, the exponential increase in training data, computations, and high user demand of AI models has led to intensive hardware resource consumption, highlighting the need to develop domain-specific semiconductors for hyperscale AI. In this technical report, we describe development trends in technologies for hyperscale AI processors pursued by domestic and foreign semiconductor companies, such as NVIDIA, Graphcore, Tesla, Google, Meta, SAPEON, FuriosaAI, and Rebellions.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송∙미디어공학회 2022년도 하계학술대회
• /
• pp.572-573
• /
• 2022

본 논문은 메타버스 환경에서 문제가 대두되고있는 AI 윤리(ethic)를 배경으로 인터랙션을 통해 사람들의 온라인과 오프라인의 결정요소에 직접적으로 영향을 미치는 대화형 AI가 어떻게 윤리적으로 진화될 수 있을지에 대한 공학적 솔루션을 UX 관점으로 찾아보는 기술 전략 연구라고 할 수 있다. 연구의 가설은 AI 의 머신러닝과정에 개별 사용자 그룹의 경험데이터가 반드시 포함되고 고려되어야 AI 는 오류값을 줄이고 윤리적으로 대응할 수 있다는 전제이다. 이를 위하여 본 논문은 기존의 머신러닝과 대화형 AI 의 UX 관점의 다이아로그 플로우 등을 연구 분석하고 사용자 데이터들을 실험하여 메타버스 서비스 환경에서의 기존에 논의되고 있는 컨택스트기반의 AI 머신러닝 과정에 사용자의 정성적 경험데이터를 추가한 윤리적 UX 접근 개념 모델을 제안 하였다. 아직은 개념모델 단계이고 시스템에서는 지금까지 다르지 않았던 비정량적인 감정과 융합적경험을 어떻게 문화적으로 코드화 하고 시스템적인 랭귀지와 연결시킬 수 있을지에 대한사용자 연구가 후속연구로 진행될 예정이다.

• 대한전자공학회논문지
• /
• 제24권3호
• /
• pp.425-432
• /
• 1987

The microprogram is the key technology of the implementation of the processor's control unit. But the coding and testing it is the most tedious process in the developing a new computer system. We developed the conversational micro architecture simulator(C-MAS) in order to use it as a microprogram development tool and a run time analyzer of the microprogram. We discuss the hardware description language(HDL) for a smulation, the design constraints of the C-MAS, and the user interface of it in this paper. We used the C language as the description language, and developed it on the berkeley UNIX4.2.

• 한국정보과학회 언어공학연구회:학술대회논문집(한글 및 한국어 정보처리)
• /
• 한국정보과학회언어공학연구회 2021년도 제33회 한글 및 한국어 정보처리 학술대회
• /
• pp.433-438
• /
• 2021

대화 시스템은 크게 사용자와 시스템이 특정 목적 혹은 자유 주제에 대해 대화를 진행하는 것으로 구분된다. 최근 자유주제 대화 시스템(Open-Domain Dialogue System)에 대한 연구가 활발히 진행됨에 따라 자유 주제를 기반으로 하는 상담 대화, 일상 대화 시스템의 독성 발화 제어 생성에 대한 연구의 중요성이 더욱 커지고 있다. 이에 본 논문에서는 대화 모델의 독성 응답 생성을 제어하기 위해 일상 대화 데이터셋으로 학습된 BART 모델에 Plug-and-Play Language Model 방법을 적용한다. 공개된 독성 대화 분류 데이터셋으로 학습된 독성 응답 분류기를 PPLM의 어트리뷰트(Attribute) 모델로 활용하여 대화 모델의 독성 응답 생성을 감소시키고 그 차이를 실험을 통해 정량적으로 비교한다. 실험 결과 어트리뷰트 모델을 활용한 모든 실험에서 독성 응답 생성이 감소함을 확인하였다.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2021년도 추계학술발표대회
• /
• pp.815-818
• /
• 2021

현재 대부분의 패션 추천시스템은 프로필 또는 설문조사를 통해 수집 된 사용자의 정적 정보를 활용하고 있다. 사용자의 정적 정보는 매우 한정적이며 이를 활용하여 다양한 환경에 적합한 패션 코디셋을 추천하기란 매우 어렵다. AI코디네이터와 사용자간의 지속적인 대화가 담긴 대화질의 데이터셋을 사용하면 사용자의 상황과 환경을 고려하여 개인에게 최적화 된 패션 코디셋을 추천할 수 있다. 본 논문에서는 한국전자통신연구원(ETRI)에서 제공하는 AI 패션 코디네이터와 사용자의 대화 정보가 담긴 FASCODE 데이터셋을 사용하여 사용자의 발화에 따라 의상을 추천하는 인공지능 모델을 위한 대화질의 데이터 전처리 방법을 제안한다.