• KIISE Transactions on Computing Practices
• /
• v.23 no.10
• /
• pp.599-604
• /
• 2017

A useful application of smart assistants is to predict and suggest users' daily behaviors the way real assistants do. Conventional methods to predict behavior have mainly used explicit schedule information logged by a user or extracted from e-mail or SNS data. However, gathering explicit information for smart assistants has limitations, and much of a user's routine behavior is not logged in the first place. In this paper, we suggest a novel approach that combines explicit schedule information with patterns of routine behavior. We propose using inference based on a Markov decision process and learning with a reward function based on inverse reinforcement learning. The results of our experiment shows that the proposed method outperforms comparable models on a life-log dataset collected over six weeks.

• The Transactions of the Korea Information Processing Society
• /
• v.13 no.6
• /
• pp.284-290
• /
• 2024

Speech emotion recognition (SER) is a technique that is used to analyze the speaker's voice patterns, including vibration, intensity, and tone, to determine their emotional state. There has been an increase in interest in artificial intelligence (AI) techniques, which are now widely used in medicine, education, industry, and the military. Nevertheless, existing researchers have attained impressive results by utilizing acted-out speech from skilled actors in a controlled environment for various scenarios. In particular, there is a mismatch between acted and spontaneous speech since acted speech includes more explicit emotional expressions than spontaneous speech. For this reason, spontaneous speech-emotion recognition remains a challenging task. This paper aims to conduct emotion recognition and improve performance using spontaneous speech data. To this end, we implement deep learning-based speech emotion recognition using the VGG (Visual Geometry Group) after converting 1-dimensional audio signals into a 2-dimensional spectrogram image. The experimental evaluations are performed on the Korean spontaneous emotional speech database from AI-Hub, consisting of 7 emotions, i.e., joy, love, anger, fear, sadness, surprise, and neutral. As a result, we achieved an average accuracy of 83.5% and 73.0% for adults and young people using a time-frequency 2-dimension spectrogram, respectively. In conclusion, our findings demonstrated that the suggested framework outperformed current state-of-the-art techniques for spontaneous speech and showed a promising performance despite the difficulty in quantifying spontaneous speech emotional expression.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2007.02a
• /
• pp.60-63
• /
• 2007

최근, 핸드폰은 일상생활의 필수품이 되었다. 특히 카메라가 장착된 핸드폰은 개인홍보와 오락요소로의 대표적인 도구로 사용된다. 나아가 양안카메라가 적용된 핸드폰은 사용자에게 입체감을 전달할 것이며, 입체감은 보다 효과적인 개인홍보 및 오락성을 제공할 것이다. 핸드폰은 휴대가 용이하지만 제한된 표시소자의 크기 및 해상도로 인하여 입체시 구현 시 사용자에게 손실된 정보 및 낮은 임장감을 제공한다. 이에 본 연구는 기존의 입체시 구현에 대한 구속조건들을 분석하고 핵심 요인을 산출하여, 핸드폰 환경에서의 최적의 입체시 구현을 위한 새로운 구속조건의 범위를 정량화하였다. 대표적인 구속조건으로는 수평, 수직 시차량이 고려되었다. 본 연구의 결과는 핸드폰용 양안카메라의 제작공차 및 효과적인 시차영상 제작을 위한 기준을 제시한다. 연구에 사용된 제품으로는 QVGA해상도(320*240) 및 2.8' 크기의 표시소자인 PDA폰이며, 입체시에 대한 구속조건을 정량화하기 위하여 수직, 수평이동이 가능한 지그를 사용하여 실험의 정확성을 높였다. 피실험자는 10명이며 남자 7명, 여자 3명으로 구성되었다.

• Journal of the Korean Society of Physical Medicine
• /
• v.3 no.4
• /
• pp.301-307
• /
• 2008

목적 : 노년층의 대부분이 가지고 있는 골관절염이라고도 알려진 퇴행성 관절 질환은 척추에 영향을 끼친다. 그리고 이 질환은 통증과 일상생활에서의 활동 제한을 초래한다. 본 연구는 퇴행성 관절질환 환자의 물리치료 빈도에 따른 효과를 검증하는 데 있다. 방법 : 퇴행성 관절 질환을 가진 30명의 환자가 이 연구에 참여하였고 이들에 취해진 물리치료의 빈도에 따른 효과를 평가하여 그 결과를 물리치료 임상 지침서와 비교분석하였다. 본 연구에 참여한 환자들은 시각적 사상척도(Visual Analog Scale, VAS), 멕길 통증 설문지, 일반적인 설문지에 응했으며 물리치료 전과 후를 물리치료 빈도에 중점을 두어 평가받았다. 결과 : 본 연구에 참여한 환자들은 제시된 치료 빈도와 평가에 근거하여 볼 때 놀랄만한 호전을 보였다. 치료 초기와 물리치료를 시작하기 전의 평가 자료를 비교하여 보았을 때 설문지로 평가된 점수에서 큰 향상을 보였다. 결론 : 본 연구는 물리치료가 퇴행성 관절 질환 환자들에게 효과가 있음을 보여주었다. 또한 물리치료 임상지침서는 따라야 할 좋은 자료가 될 수 있지만 물리치료 처방에는 정해진 기준은 없다는 것을 보여주었다.

• Resources Recycling
• /
• v.12 no.4
• /
• pp.44-50
• /
• 2003

Separation of mixed plastics is not a simple task due to similarities of specific gravity and limitations of sizes, etc. In this study, a new technique was developed to separate plastics selectively by a thermal adhesion method using the different softening or melting temperatures of various types of plastics. Tests were conducted to investigate softening or melting temperatures and adhesion probabilities of plastics with change in temperature. Tests also were conducted to analyze the operating variables, including contact times, size and color pigment. It was founded that LDPE, HDPE and PVC, PET, etc among mixed plastics could be separated selectively according to the change of temperature.

• The Korean Journal of Health Service Management
• /
• v.12 no.1
• /
• pp.95-107
• /
• 2018

Objectives : This study was performed to investigate the health status and social support among elderly living alone with restricted daily functions using the data of the "2014 the Korean Elderly Survey". Methods : Data on 2,407 elderly living alone were drawn and statistically examined using a t-test, an ANOVA, and a multiple regression analysis. Results : The study found that first, the elderly living alone with restricted daily functions comprised 22.1% of the total elderly living alone, and those who were older elderly, illiterate, with low-income, having poor nutrition management, and with a poor health status. Second, among the elderly living alone with restricted daily functions, there was a group with very little support from the family and only 14.0% were covered by long-term care insurance. Third, the life satisfaction of the elderly with restricted functions was lower than that of the non-restricted elderly, and was affected by income, health conditions, depression, access to senior welfare centers, and communication with others. Conclusions : The elderly living alone with restricted daily functions have serious health risks and social support, and hence they should be provided with more proactive support for life, health care and social care to live independently within their communities.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2017.11a
• /
• pp.764-767
• /
• 2017

최근 개인정보 활용에서 보안 및 인증에 대한 중요성이 대두되고 있으며, IoT 기반의 바이오 디바이스를 접목한 생체인식 기술은 사용자의 식별과 인증을 위해 다양한 분야에서 많은 발전을 보이고 있다. 본 논문에서는 대규모 뇌파신호를 효과적으로 처리하기 위한 방안으로 Deep Learning 기법을 적용한 뇌파 데이터 식별과 이를 분석한 사용자 인증이 가능한 스마트 자물쇠 모델을 제안한다. 제안한 뇌파신호를 활용한 Deep Learning 기반 사용자 식별 및 인증 모델은 보안 시스템에서의 활용뿐만 아니라 다양한 사물인터넷과 접목 시킬 수 있으며, 뇌성마비 또는 신체 활동이 제한적인 환자의 경우 일상생활의 제약을 줄이고 삶의 질적 향상에 도움이 될 것으로 기대한다.

• Journal of the Korea Convergence Society
• /
• v.10 no.4
• /
• pp.147-151
• /
• 2019

Modern people spend most of their daily lives in their homes, schools, or workplaces, hospitals, shopping malls, subway stations, rooms, and parking lots. According to the survey, air quality management at the multi-use facility is less than 50% satisfied. In this study, a photocatalytic filtration system is developed by utilizing a broadband-to-active photocatalyst that utilizes a media photocatalyst filter that removes airborne germs from indoor air as well as indoor air quality and operates on visible light as well as ultraviolet light.

• The Journal of Korean society of community based occupational therapy
• /
• v.9 no.2
• /
• pp.1-12
• /
• 2019

Objective : The purpose of this study is to measure the level of activity possessed by the daily activities of the elderly. In addition, we attempted to compare the differences in the categories of daily activities. Methods : We used K-ACS to evaluate 30 older people who engaged in economic activity living in community, 30 older people who did not have economic activity living in community, and 30 older people living in facility. We compared the level of activity retention. Results : There was a significant difference of level of activity retention according to the daily activity category of the subjects(p <.001). There was a difference in the instrumental activities according to the category of each activity, while the three groups showed similar patterns in leisure activities and social activities. In particular, all three groups were most likely to rest and watch television in the leisure activities, so the leisure activities of the elderly in Korea were very limited. Among the general characteristics, it was analyzed that there was a correlation between the activity, the marital status, the religion and the cognitive level. Conclusion : According to the daily activities category of the elderly, there was a difference in level of activity retention and the characteristics of daily activities correlated with the activities of the elderly. This suggests that it may be a basic data for establishing a client - centered intervention plan by considering the characteristics affecting daily living and level of activity retention of the elderly in the occupational therapy.

• Therapeutic Science for Rehabilitation
• /
• v.2 no.2
• /
• pp.5-20
• /
• 2013

Constraint Induced Movement Therapy(CIMT) is intense in that patient's unaffected arm is restrained for 90% of waking hours during a two-week period while they also participate in activity sessions using the affected arm for 6 hours/day. However CIMT showed that an issue for applying it to clinics of patients with stroke, and then modified constraint induced movement therapy(mCIMT) was designed to minimize the issue. Application on mCIMT for the patients has been studied in various ways. As a result, it has proved the effect on functional improvement of patients with stroke through methods such as MAL, FMA, WMFT, ARAT, FIM, SIS and so forth. It's considered that modified constraint induced movement therapy can be useful applied on clinical experiments of occupational therapy, as it is a way of treatment of upper extremity function, activities of daily living and an improvement of the quality of life for stroke patients.