• Annual Conference of KIPS
• /
• 2019.10a
• /
• pp.830-832
• /
• 2019

IoT 기술의 도래로 인하여 실생활에 사용되는 사물들에 Sensor가 부착되어 시간마다 Sensor data가 발생하는 세상이 열리게 되었다. 이러한 IoT Device들에 부착되어 있는 sensor를 통하여 수집이 된 data는 방대한 양을 가지기 때문에 Deep Learning에 적용하는데 충분하며 아주 중요한 역할을 한다. 이러한 IoT Device들은 우리의 실제 생활에 아주 가까이 다양한 환경으로 접할 수 있다. 예를 들어 스마트시티, 스마트팩토리, 스마트홈 등이 있다. 이러한 것들은 우리의 일상생활에 편리함과 직결되어 있다. 본 논문에서는 Smart home 환경에서의 Multi Resident Activity Recognition이다. Smart home의 가구에 부착되어 있는 센서에서 발생된 센서데이터를 활용하여 1) Training Similarity Network, 2) Embedding, 3) Clustering, 4) Recognizing 네 단계 프로세스를 거쳐 문제를 해결한다. 그 결과, 우리가 제안한 프로세스를 통하여 차원 축소 효과와 Un-seen data를 효과적으로 처리할수 있게 된다.

• Journal of KIISE:Computing Practices and Letters
• /
• v.15 no.4
• /
• pp.260-264
• /
• 2009

There have been many studies on u-Healthcare system for human using sensor network systems. In this paper, we design and implement a healthcare system for pets called uPetCare(Ubiquitous Pet-Care System) that can manage the status of pet on the web. The main functions of this system are 1) gathering data using sensor network, 2) multi-hop communication in sensor network, 3) data compression and aggregation at sink node, 4) storing data in web server, 5) real-time data monitoring using AJAX, 6) activity recognition of pet.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.18 no.6
• /
• pp.319-325
• /
• 2023

In this study, we proposed a method for hyperparameter optimization in the building and training of a deep learning model designed to process point cloud data collected by a millimeter-wave radar system. The primary aim of this study is to facilitate the deployment of a baseline model in resource-constrained IoT devices. We evaluated a RadHAR baseline deep learning model trained on a public dataset composed of point clouds representing five distinct human activities. Additionally, we introduced a coarse-to-fine hyperparameter optimization procedure, showing substantial potential to enhance model efficiency without compromising predictive performance. Experimental results show the feasibility of significantly reducing model size without adversely impacting performance. Specifically, the optimized model demonstrated a 3.3% improvement in classification accuracy despite a 16.8% reduction in number of parameters compared th the baseline model. In conclusion, this research offers valuable insights for the development of deep learning models for resource-constrained IoT devices, underscoring the potential of hyperparameter optimization and model size reduction strategies. This work contributes to enhancing the practicality and usability of deep learning models in real-world environments, where high levels of accuracy and efficiency in data processing and classification tasks are required.

• Journal of Engineering Education Research
• /
• v.9 no.3
• /
• pp.37-48
• /
• 2006

The purpose of this study is to investigate the response and the recognition of students of blended learning. Set up the following study problem in order to achieve a study purpose. How are a response and a recognition of learner about blended learning? To study, Designed blended learning environment, and analyzed a bulletin board and a log file, and made up a question for 15 weeks after proceeding a class in order to recognition and response about the results operating blended learning. The results as follows: First, A lot of students prefer education in blended learning environment, and participate positively in activity in on-line activity. Second, When study in blended learning environment about media be prudent must select. Third, Confirmed the fact that various role of Instructor needed. Therefore, the design and operation of blended learning, which mixes traditional teaching methods and a modern on-line method, is meaningful.

• Journal of Preventive Medicine and Public Health
• /
• v.50 no.5
• /
• pp.320-327
• /
• 2017

Objectives: The purpose of our study was to evaluate the association between the intensity of physical activity (PA) and prevalence of cardiovascular disease (CVD) using Korean representative data. Methods: We analyzed 39 804 participant data from the Korea National Health and Nutrition Examination Survey, 2007-2013. Exposure variable was three levels of PA (low, medium, and high) in a week, and outcome variable was prevalence of CVD based on patient self-recognition and doctor's diagnosis. Complex logistic regression analysis was performed to evaluate the relationship between level of PA and CVD adjusted by body mass index, hypertension, hypercholesterolemia, diabetes mellitus, stress recognition, household income, smoking, and current drinking. The indices of association w ere estimated as crude prevalence odds ratio (POR), adjusted POR, and their 95% confidence interval (CI). All statistical analyzes were performed using complex sample analysis procedure of the SPSS version 23.0. Results: When all variables were adjusted, only high level PA in women showed a significant association with stroke (adjusted POR by patient's self-recognition, 0.57; 95% CI, 0.32 to 0.99, adjusted POR by doctor's diagnosis, 0.55; 95% CI, 0.34 to 0.87) and CVD (adjusted POR by doctor's diagnosis, 0.68; 95% CI, 0.48 to 0.96). Conclusions: High level PA in women has a significant reverse association with prevalence of stroke and CVD in Korea. Further study for elucidating the mechanism will be needed.

• Korean Journal of Food Science and Technology
• /
• v.27 no.5
• /
• pp.646-651
• /
• 1995

The correlation between the bitter taste and the surface tension was found for bitter tasting, aqueous solutions. By the Szyszkowski's equation, the surface tension (STR) and taste curves (JND) were derived more clearly using the Techplot program. The specific capillary activity (log b values) for bitter tasting solutions are negatively correlated to the recognition threshold. It was shown that a more bitter substances has greater capillary activity. The correlation between the recognition threshold $(log\;C_{1})$ and the substance specific constant (a and b values) of sensory (JND) and surface tension values indicates good agreement. This means that the model of surface area adsorption in the solution/air system can apply also for the sensory model microvillus membrane in the mouth.

• Molecules and Cells
• /
• v.41 no.6
• /
• pp.591-602
• /
• 2018

Gastric cancer is the fifth most common type of malignancy worldwide, and the survival rate of patients with advanced-stage gastric cancer is low, even after receiving chemotherapy. Here, we validated neurotensin receptor 1 (NTSR1) as a potential therapeutic target in gastric cancer. We compared NTSR1 expression levels in sixty different gastric cancer-tissue samples and cells, as well as in other cancer cells (lung, breast, pancreatic, and colon), by assessing NTSR1 expression via semi-quantitative real-time reverse transcription polymerase chain reaction, immunocytochemistry and western blot. Following neurotensin (NT) treatment, we analyzed the expression and activity of matrix metalloproteinase-9 (MMP-9) and further determined the effects on cell migration and invasion via wound-healing and transwell assays. Our results revealed that NTSR1 mRNA levels were higher in gastric cancer tissues than non-cancerous tissues. Both of NTSR1 mRNA levels and expression were higher in gastric cancer cell lines relative to levels observed in other cancer-cell lines. Moreover, NT treatment induced MMP-9 expression and activity in all cancer cell lines, which was significantly decreased following treatment with the NTSR1 antagonist SR48692 or small-interfering RNA targeting NTSR1. Furthermore, NT-mediated metastases was confirmed by observing epithelial-mesenchymal transition markers SNAIL and E-cadherin in gastric cancer cells. NT-mediated invasion and migration of gastric cancer cells were reduced by NTSR1 depletion through the Erk signaling. These findings strongly suggested that NTR1 constitutes a potential therapeutic target for the inhibition of gastric cancer invasion and metastasis.

• Journal of the Korea Convergence Society
• /
• v.11 no.9
• /
• pp.73-81
• /
• 2020

The purpose of this study was to analyze the recognition of 20s~30s' young people in marathon activities using Q methodology. P-sample was composed of 30 young marathoners. The selected Q samples were arranged in the normal distribution form. The collected data were analyzed by factor analysis through varimax rotation using QUANL 1.2PC program. Three distinct were emerged throughout the analysis: the first type is 'the type of giving special meaning.' which is improving their physical fitness, healthy and enjoying runner's high. The second type is 'the enjoying leisure' which is releasing stress and enjoying their common interest.' The last type is 'the type of enjoying crew activity.' which is enjoying marathon festival and managed their healthy and life. This result hints that young marathoners' sports culture trends.

• Management & Information Systems Review
• /
• v.32 no.2
• /
• pp.59-81
• /
• 2013

Nowadays, social responsibility activities's influence on company's production activity is expanding and corporate social responsibility as sustainability is mandatory. Unlike the past that few companies did social responsibility activity, it is hard to find differences in present that lots of companies are doing social responsibility activity. This research divides social group and corporate group according to the consumer's recognition about company's social responsibility activity and affect-relation of consumer's recognition is confirmed based on the level of authenticity when company does social responsibility activity. As a result, this research states that company's social responsibility activity will make conclusive responsibility on increasing consumer's trust on the company by letting consumers recognize that company's social responsibility activity whiling achieving its seeking goal is not for company's profit, but is sincere action toward social, which means it is the responsibility activity including authenticity.

• KIPS Transactions on Computer and Communication Systems
• /
• v.7 no.8
• /
• pp.211-218
• /
• 2018

In this study, we propose an algorithm to detect eating activity and estimation mealtime using 6-axis inertial sensor. The eating activity is classified into three types: food picking, food eating, and lowering. The feature points of the gyro signal are selected for each gesture, and the eating activity is detected when each feature point appears in the sequence. Morphology technique is used to post-process to detect meal time. The proposed algorithm achieves the accuracy of 94.3% and accuracy of 84.1%.