• Annual Conference on Human and Language Technology
• /
• 2019.10a
• /
• pp.78-80
• /
• 2019

개체명 인식 연구에서 우수한 성능을 보이고 있는 bi-LSTM-CRFs 모델은 처리 속도가 느린 단점이 있고, CNN-CRFs 모델은 한국어 말뭉치를 사용하여 제대로 분석되지 않았다. 본 논문에서는 한국어 개체명 인식 말뭉치를 이용한 CNN-CRFs 모델의 음절 단위 한국어 개체명 인식 방법을 제안한다. 실험 결과 bi-LSTM-CRFs 모델보다 CNN-CRFs 모델의 F1 score가 0.4% 높았고, 27.5% 빠른 처리 속도를 보였다.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.8
• /
• pp.23-30
• /
• 2022

Deep learning analyzes data to discover a series of rules and anticipates the future, helping us in various ways in our lives. For example, prediction of stock prices and agricultural prices. In this research, the results of solar photovoltaic power generation accompanied by weather are analyzed through deep learning in situations where the importance of solar energy use increases, and the amount of power generation is predicted. In this research, we propose a model using LSTM(Long Short Term Memory network) that stand out in time series data prediction. And we compare LSTM's performance with CNN(Convolutional Neural Network), which is used to analyze various dimensions of data, including images, and CNN-LSTM, which combines the two models. The performance of the three models was compared by calculating the MSE, RMSE, R-Squared with the actual value of the solar photovoltaic power generation performance and the predicted value. As a result, it was found that the performance of the LSTM model was the best. Therefor, this research proposes predicting solar photovoltaic power generation using LSTM.

• 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.

• New & Renewable Energy
• /
• v.18 no.2
• /
• pp.18-25
• /
• 2022

In this study, we propose a wind power generation prediction system that applies machine learning and data mining to predict wind power generation. This system increases the utilization rate of new and renewable energy sources. For time-series data, the data set was established by measuring wind speed, wind generation, and environmental factors influencing the wind speed. The data set was pre-processed so that it could be applied appropriately to the model. The prediction system applied the CNN (Convolutional Neural Network) to the data mining process and then used the LSTM (Long Short-Term Memory) to learn and make predictions. The preciseness of the proposed system is verified by comparing the prediction data with the actual data, according to the presence or absence of data mining in the model of the prediction system.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.11a
• /
• pp.939-941
• /
• 2021

사람은 표정, 음성, 말 등을 통해 감정을 표출한다. 본 논문에서는 화자의 음성데이터만을 사용하여 감정을 분류하는 방법을 제안한다. 멜 스펙트로그램(Mel-Spectrogram)을 이용하여 음성데이터를 시간에 따른 주파수 영역으로 변화한다. 멜 스펙트로그램으로 변환된 데이터를 CNN을 이용하여 특징 벡터화한 후 Bi-Directional LSTM을 이용하여 화자의 발화 시간 동안 변화되는 감정을 분석한다. 마지막으로 완전 연결 네트워크를 통해 전체 감정을 분류한다. 감정은 Anger, Excitement, Fear, Happiness, Sadness, Neutral로, 총 6가지로 분류하였으며 데이터베이스로는 상명대 연구팀에서 구축한 한국어 음성 감정 데이터베이스를 사용하였다. 실험 결과 논문에서 제안한 CNN-LSTM 모델의 정확도는 88.89%로 측정되었다.

• Journal of Information Processing Systems
• /
• v.20 no.2
• /
• pp.159-172
• /
• 2024

The number of healthcare products available for pets has increased in recent times, which has prompted active research into wearable devices for pets. However, the data collected through such devices are limited by outliers and missing values owing to the anomalous and irregular characteristics of pets. Hence, we propose pet behavior recognition based on a hybrid one-dimensional convolutional neural network (CNN) and long short- term memory (LSTM) model using pet wearable devices. An Arduino-based pet wearable device was first fabricated to collect data for behavior recognition, where gyroscope and accelerometer values were collected using the device. Then, data augmentation was performed after replacing any missing values and outliers via preprocessing. At this time, the behaviors were classified into five types. To prevent bias from specific actions in the data augmentation, the number of datasets was compared and balanced, and CNN-LSTM-based deep learning was performed. The five subdivided behaviors and overall performance were then evaluated, and the overall accuracy of behavior recognition was found to be about 88.76%.

• Transactions of Materials Processing
• /
• v.33 no.3
• /
• pp.214-230
• /
• 2024

• 한국어정보학회:학술대회논문집
• /
• 2017.10a
• /
• pp.82-84
• /
• 2017

본 논문에서는 한국어 이미지 캡션을 학습하기 위한 데이터를 작성하고 딥러닝을 통해 예측하는 모델을 제안한다. 한국어 데이터 생성을 위해 MS COCO 영어 캡션을 번역하여 한국어로 변환하고 수정하였다. 이미지 캡션 생성을 위한 모델은 CNN을 이용하여 이미지를 512차원의 자질로 인코딩한다. 인코딩된 자질을 LSTM의 입력으로 사용하여 캡션을 생성하였다. 생성된 한국어 MS COCO 데이터에 대해 어절 단위, 형태소 단위, 의미형태소 단위 실험을 진행하였고 그 중 가장 높은 성능을 보인 형태소 단위 모델을 영어 모델과 비교하여 영어 모델과 비슷한 성능을 얻음을 증명하였다.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.20 no.1
• /
• pp.86-99
• /
• 2021

One of the methods to alleviate traffic congestion is to increase the efficiency of the roads by providing traffic condition information on road user and distributing the traffic. For this, reliability must be guaranteed, and quantitative real-time traffic speed prediction is essential. In this study, and based on analysis of traffic speed related to traffic conditions, historical data correlated with traffic flow were used as input. We developed an LSTM model that predicts speed in response to normal traffic conditions, along with a CNN-LSTM model that predicts speed in response to incidents. Through these models, we try to predict traffic speeds during the hour in five-minute intervals. As a result, predictions had an average error rate of 7.43km/h for normal traffic flows, and an error rate of 7.66km/h for traffic incident flows when there was an incident.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.11
• /
• pp.1471-1476
• /
• 2021

Flood damage can cause floods or tsunamis, which can result in enormous loss of life and property. In this regard, damage can be reduced by making a quick evacuation decision through flood prediction, and many studies are underway in this field to predict floods using time series data. In this paper, we propose a CNN-based time series prediction model. A CNN-based water level prediction model was implemented using the river level and precipitation, and the performance was confirmed by comparing it with the LSTM and GRU models, which are often used for time series prediction. In addition, by checking the performance difference according to the size of the input data, it was possible to find the points to be supplemented, and it was confirmed that better performance than LSTM and GRU could be obtained. Through this, it is thought that it can be utilized as an initial study for flood prediction.