• Journal of Korean Society of Water and Wastewater
• /
• v.35 no.1
• /
• pp.83-91
• /
• 2021

The increased turbidity in rivers during flood events has various effects on water environmental management, including drinking water supply systems. Thus, prediction of turbid water is essential for water environmental management. Recently, various advanced machine learning algorithms have been increasingly used in water environmental management. Ensemble machine learning algorithms such as random forest (RF) and gradient boosting decision tree (GBDT) are some of the most popular machine learning algorithms used for water environmental management, along with deep learning algorithms such as recurrent neural networks. In this study GBDT, an ensemble machine learning algorithm, and gated recurrent unit (GRU), a recurrent neural networks algorithm, are used for model development to predict turbidity in a river. The observation frequencies of input data used for the model were 2, 4, 8, 24, 48, 120 and 168 h. The root-mean-square error-observations standard deviation ratio (RSR) of GRU and GBDT ranges between 0.182~0.766 and 0.400~0.683, respectively. Both models show similar prediction accuracy with RSR of 0.682 for GRU and 0.683 for GBDT. The GRU shows better prediction accuracy when the observation frequency is relatively short (i.e., 2, 4, and 8 h) where GBDT shows better prediction accuracy when the observation frequency is relatively long (i.e. 48, 120, 160 h). The results suggest that the characteristics of input data should be considered to develop an appropriate model to predict turbidity.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.10a
• /
• pp.950-951
• /
• 2019

시계열 데이터에 대한 군집화 관련 연구는 주로 통계 분석을 통해 이뤄지기 때문에 데이터가 갖는 특성을 완전히 반영하는 데 한계를 갖는다. 본 논문에서는 다변량 데이터에서의 군집화를 위하여 변수별로 시간에 따른 변화와 특징을 추출하기 위한 CNN-GRU(Convolutional Neural Network - Gated Recurrent Unit) 기반의 신경망 모델을 제안한다. CNN을 활용하여 변수별로 갖는 특성을 파악하고자 하였으며, GRU을 통해 전체 시간에 따른 소비 추세를 도출하고자 하였다. 지역별로 업종에 따라 사용된 2년 치의 실제 카드 데이터를 활용하였으며, 유사한 소비 추세를 보이는 지역을 군집화하는데 이를 적용하였다. 결과적으로, 다변량 시계열 데이터를 통해 전체적인 흐름을 반영하여 패턴화했다는 점에서 의의를 갖는다.

• Journal of the Korea Society of Computer and Information
• /
• v.23 no.11
• /
• pp.31-41
• /
• 2018

Text classification has been studied for a long time in the Natural Language Processing field. In this paper, we propose an article- and paragraph-level genre classification system using Word2Vec-based LSTM, GRU, and CNN models for large-scale English corpora. Both article- and paragraph-level classification performed best in accuracy with LSTM, which was followed by GRU and CNN in accuracy performance. Thus, it is to be confirmed that in evaluating the classification performance of LSTM, GRU, and CNN, the word sequential information for articles is better than the word feature extraction for paragraphs when the pre-trained Word2Vec-based word embeddings are used in both deep learning-based article- and paragraph-level classification tasks.

• Communications for Statistical Applications and Methods
• /
• v.28 no.4
• /
• pp.329-338
• /
• 2021

Deep learning methods have been developed, used in various fields, and they have shown outstanding performances in many cases. Many studies predicted a daily stock return, a classic example of time-series data, using deep learning methods. We also tried to apply deep learning methods to Korea's stock market data. We used Korea's stock market index (KOSPI) and several individual stocks to forecast daily returns and directions. We compared several deep learning models with other machine learning methods, including random forest and XGBoost. In regression, long short term memory (LSTM) and gated recurrent unit (GRU) models are better than other prediction models. For the classification applications, there is no clear winner. However, even the best deep learning models cannot predict significantly better than the simple base model. We believe that it is challenging to predict daily stock return data even if we use the latest deep learning methods.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.612-614
• /
• 2022

최근 딥러닝 기술은 자연어처리에서 기본적이고 필수적인 기법으로 자연어처리에 필요한 복잡한 비선형 관계를 모델링할 수 있다. 본 논문에서는 LSTM(Long Short-Term Memory)과 GRU(Gated Recurrent Unit) 딥러닝 기술을 연구 논문 분류에 적용하며, CNN(Convolutional Neural Network)에 LSTM과 GRU을 각각 결합하여 특정 분야의 연구 논문을 분류하고 연구 논문을 추천하는 기법을 제안한다. 워드 임베딩과 딥러닝 기법을 연구 논문 분류에 적용하여 관심이 있는 단어와 단어 주변의 단어들 사이의 유사성과 성능을 비교 분석한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.287-288
• /
• 2023

우리 사회가 고령화시대로 접어들면서 낙상은 매우 심각한 사회문제가 되고 있으며 정확한 낙상 감지 기술의 수요도 늘고 있다. 본 연구는 웹 캠을 이용한 개선된 낙상감지 기법을 제안한다. 제안하는 기법은 RGB 영상을 기반으로 스켈레톤 포즈 추출, 데이터 가공, GRU(Gated Recurrent Unit) 신경망 알고리즘을 적용한 낙상 감지 실험 및 감지 결과 분석의 과정이 포함된다.

• Korean Journal of Applied Biomechanics
• /
• v.34 no.2
• /
• pp.71-80
• /
• 2024

Objective: This study aimed to explore the brain connectivity between brain and biomechanical variables by exploring motion recognition through FFT (fast fourier transform) analysis and AI (artificial intelligence) focusing on quiet standing movement patterns. Method: Participants included 12 young adult males, comprising university students (n=6) and elite gymnasts (n=6). The first experiment involved FFT of biomechanical signals (f_CoP, f_AJtorque and f_EEG), and the second experiment explored the optimization of AI-based GRU (gated recurrent unit) using f_EEG data. Results: Significant differences (p<.05) were observed in frequency bands and maximum power based on group and posture types in the first experiment. The second study improved motion prediction accuracy through GRU performance metrics derived from brain signals. Conclusion: This study delved into the movement pattern of upright standing posture through the analysis of bio-signals linking the cerebral cortex to motor performance, culminating in the attainment of motion recognition prediction performance.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2024.05a
• /
• pp.37-40
• /
• 2024

본 연구에서는 웨어러블 디바이스로부터 수집된 다변량 반려동물 행동 데이터를 처리하기 위해, GCN(Graph Convolutional Network)과 GRU(Gated Recurrent Unit)를 결합한 모델을 제안한다. 제안된 모델은 시계열 내부 구조를 활용하여 그래프 구조로 변환하고, DTW(Dynamic Time Warping) 유사도 분석을 통해 노드 간의 시간적 유사도를 기반으로 엣지를 생성한다. 실험결과로 DTW 기반 엣지 생성 방식이 유클리드 거리 및 선형 방식에 비해 더 높은 성능을 나타냈다. 본 연구는 반려동물의 행동을 정확히 분류하기 위한 효과적인 방법론을 제공한다.

• KIPS Transactions on Software and Data Engineering
• /
• v.6 no.9
• /
• pp.457-464
• /
• 2017

Indoor air quality analysis is conducted to understand abnormal atmospheric phenomena and the external factor affecting indoor air quality. By recording indoor air quality measurements periodically, we are able to observe patterns in air quality. However, it difficult to predict the number of potential parameters, set parameters for a given observation and find the coefficients. Moreover, the results are time-dependent. Thus to address these issues, we introduce a microchip capable of periodically recording indoor air quality and a model that estimates atmospheric changes based on time series data.

• Journal of Korea Multimedia Society
• /
• v.25 no.9
• /
• pp.1257-1265
• /
• 2022

Recognition of sign language is challenging due to the occlusion of hands, accuracy of hand gestures, and high computational costs. In recent years, deep learning techniques have made significant advances in this field. Although these methods are larger and more complex, they cannot manage long-term sequential data and lack the ability to capture useful information through efficient information processing with faster convergence. In order to overcome these challenges, we propose a word-level sign language recognition (SLR) system that combines a real-time human pose detection library with the minimized version of the gated recurrent unit (GRU) model. Each gate unit is optimized by discarding the depth-weighted reset gate in GRU cells and considering only current input. Furthermore, we use sigmoid rather than hyperbolic tangent activation in standard GRUs due to performance loss associated with the former in deeper networks. Experimental results demonstrate that our pose-based optimized GRU (Pose-OGRU) outperforms the standard GRU model in terms of prediction accuracy, convergency, and information processing capability.