A5078608683- Advanced Chemical Sensor Technologies
- Gas Sensing Nanomaterials and Sensors
- IoT and Edge/Fog Computing
- Energy Efficient Wireless Sensor Networks
- Machine Learning in Healthcare
- Topic Modeling
- Artificial Intelligence in Healthcare
- Cloud Computing and Resource Management
- Water Quality Monitoring and Analysis
- Analytical Chemistry and Sensors
- Context-Aware Activity Recognition Systems
- Water Quality Monitoring Technologies
- Anomaly Detection Techniques and Applications
- Distributed and Parallel Computing Systems
- Insect Pheromone Research and Control
- Peer-to-Peer Network Technologies
- Video Surveillance and Tracking Methods
Electronics and Telecommunications Research Institute
2016-2023
Detecting illegal drugs, such as cannabis and methamphetamine, with high accuracy speed is a complex problem that requires an innovative solution. To address this challenge, we propose new method utilizes newly developed electronic nose (e-nose) system unprecedented total of 56 sensors, including four different types: metal-oxide-semiconductor (MOS), electrochemical (EC), non-dispersive infrared (NDIR), photoionization detector (PID). Previous studies on gas sensors have typically validated...
There are various medical features associated with cardiovascular disease in the EMR data, but frequency of each feature is different. Less frequent may be considered as non-critical feature, although closely risk prediction model. We propose a frequency-aware based Attention-based LSTM (FA-Attn-LSTM) that weighs on important using an attention mechanism considers feature. Our model predicts for ejection fraction target and shows RMSE = 3.65 MAE 2.49.
Gas classification is a machine learning problem that important for various applications including monitoring systems, health care, public security, etc. Since measuring the characteristic of gas molecules greatly affected by external factors such as wind speed and internal setting detecting sensors, should be done taking into account combination these individual factors, which we call <i>condition</i> in this paper. In particular, when classifying data measured under multiple conditions,...
In this paper, we proposed a hierarchical analysis architecture for IoT streaming data by switching computing load between an server and edge devices dynamically in order to reduce the resource usage of server. The hierarchically distributed system can analyze faster more with same capability. To show feasibility method, applied method prediction problem energy consumption simulated BEMS (Building Energy Management System) which is based on real data.
Detecting illegal drugs, such as cannabis and methamphetamine, with high accuracy speed is a complex problem that requires an innovative solution. To address this challenge, we propose new method utilizes newly developed electronic nose (e-nose) system unprecedented total of 56 sensors, including four different types: metal-oxide-semiconductor (MOS), electrochemical (EC), non-dispersive infrared (NDIR), photoionization detector (PID). Previous studies on gas sensors have typically validated...
The Deep Nose project aims to develop an intelligent olfactory system using gas sensors and AI for efficient inspection of smuggling items, particularly narcotics. This study introduces the System, which incorporates a sensor array consisting 56 different types achieve multimodality multidimensionality. collects data by injecting target gases mixed with various background utilizes algorithm analysis. Preliminary results show promising detection accuracy narcotics other gases. Ongoing...
We introduce the LSTM-MDN-ATTN model for predicting medical time-series data. The predicts future value of data by approximating distribution target Since is multivariate with various test items, attention mechanism used to suitable layer in this study focusing on that related proposed shows better results compared baseline models using lab from Asan Medical Center Seoul.