• 제목/요약/키워드: DNN 모델

검색결과 192건 처리시간 0.023초

Spatio-Temporal Incidence Modeling and Prediction of the Vector-Borne Disease Using an Ecological Model and Deep Neural Network for Climate Change Adaption (기후 변화 적응을 위한 벡터매개질병의 생태 모델 및 심층 인공 신경망 기반 공간-시간적 발병 모델링 및 예측)

  • Kim, SangYoun;Nam, KiJeon;Heo, SungKu;Lee, SunJung;Choi, JiHun;Park, JunKyu;Yoo, ChangKyoo
    • Korean Chemical Engineering Research
    • /
    • 제58권2호
    • /
    • pp.197-208
    • /
    • 2020
  • This study was carried out to analyze spatial and temporal incidence characteristics of scrub typhus and predict the future incidence of scrub typhus since the incidences of scrub typhus have been rapidly increased among vector-borne diseases. A maximum entropy (MaxEnt) ecological model was implemented to predict spatial distribution and incidence rate of scrub typhus using spatial data sets on environmental and social variables. Additionally, relationships between the incidence of scrub typhus and critical spatial data were analyzed. Elevation and temperature were analyzed as dominant spatial factors which influenced the growth environment of Leptotrombidium scutellare (L. scutellare) which is the primary vector of scrub typhus. A temporal number of diseases by scrub typhus was predicted by a deep neural network (DNN). The model considered the time-lagged effect of scrub typhus. The DNN-based prediction model showed that temperature, precipitation, and humidity in summer had significant influence factors on the activity of L. scutellare and the number of diseases at fall. Moreover, the DNN-based prediction model had superior performance compared to a conventional statistical prediction model. Finally, the spatial and temporal models were used under climate change scenario. The future characteristics of scrub typhus showed that the maximum incidence rate would increase by 8%, areas of the high potential of incidence rate would increase by 9%, and disease occurrence duration would expand by 2 months. The results would contribute to the disease management and prediction for the health of residents in terms of public health.

Radiation Prediction Based on Multi Deep Learning Model Using Weather Data and Weather Satellites Image (기상 데이터와 기상 위성 영상을 이용한 다중 딥러닝 모델 기반 일사량 예측)

  • Jae-Jung Kim;Yong-Hun You;Chang-Bok Kim
    • Journal of Advanced Navigation Technology
    • /
    • 제25권6호
    • /
    • pp.569-575
    • /
    • 2021
  • Deep learning shows differences in prediction performance depending on data quality and model. This study uses various input data and multiple deep learning models to build an optimal deep learning model for predicting solar radiation, which has the most influence on power generation prediction. did. As the input data, the weather data of the Korea Meteorological Administration and the clairvoyant meteorological image were used by segmenting the image of the Korea Meteorological Agency. , comparative evaluation, and predicting solar radiation by constructing multiple deep learning models connecting the models with the best error rate in each model. As an experimental result, the RMSE of model A, which is a multiple deep learning model, was 0.0637, the RMSE of model B was 0.07062, and the RMSE of model C was 0.06052, so the error rate of model A and model C was better than that of a single model. In this study, the model that connected two or more models through experiments showed improved prediction rates and stable learning results.

Particular Matter Concentration Prediction Models Based on EEMD (EEMD 기반의 미세먼지 농도 예측 모델)

  • Jung, Yong-jin;Lee, Jong-sung;Oh, Chang-heon
    • Proceedings of the Korean Institute of Information and Commucation Sciences Conference
    • /
    • 한국정보통신학회 2021년도 추계학술대회
    • /
    • pp.345-347
    • /
    • 2021
  • Various studies are being conducted to improve the accuracy of fine dust, but there is a problem that deep learning models are not well learned due to various characteristics according to the concentration of fine dust. This paper proposes an EEMD-based fine dust concentration prediction model to decompose the characteristics of fine dust concentration and reflect the characteristics. After decomposing the fine dust concentration through EEMD, the final fine dust concentration value is derived by ensemble of the prediction results according to the characteristics derived from each. As a result of the model's performance evaluation, 91.7% of the fine dust concentration prediction accuracy was confirmed.

  • PDF

Potential of LUT-based PIM for DNNs (DNN 모델 추론을 위한 Lookup Table 기반 PIM의 가능성)

  • Junguk Hong;Jinho Lee
    • The Transactions of the Korea Information Processing Society
    • /
    • 제13권11호
    • /
    • pp.590-596
    • /
    • 2024
  • Processing-in-memory (PIM) is an accelerator that enables data to be processed closer to the stored memory. Due to the nature of PIM specialized in data movement, PIM can be used as an efficient accelerator in transformer-based generative model and recommendation system, which have recently garnered attention. In this paper, we examine the latest research trends of PIM that enhance the recommendtation system and generative language model, which are suitable applications for PIM usage. Additionally, we discuss the research direction of PIM based on previous studies. Lastly, we verify the effectiveness of LUT in PIM systems through experiments.

Study on Prediction of Attendance Using Machine Learning (머신러닝을 이용한 관중 수요 예측에 관한 연구)

  • Yoo, Ji-Hyun
    • Journal of IKEEE
    • /
    • 제23권4호
    • /
    • pp.1243-1249
    • /
    • 2019
  • People who gathered to enjoy a specific event or content are called audiences or spectators, and show various propensity according to the characteristics of the crowd. Although there is such a difference, in general, the number of attendance is directly related to the business aspect, which enables stable financial operation for the sale of contents through various incomes, such as the admission fee and the use of other facilities. Therefore, prediction of audience can be used as a major factor in marketing and budgeting strategies. In this study, we review several existing models for predicting the number of attendance and propose an efficient machine learning model. In addition, we studied daily attendance prediction and abnormal attendance prediction using combine DNN(Deep Neural Network) and RF(Random Forest) model.

A Learning Rate Model of Deep Learning for Classification Analysis of Problematic Smartphone Use (스마트폰 과의존 분류 분석을 위한 딥러닝 학습률 모델)

  • Kim, Yu Jeong;Lee, Dong Su
    • Proceedings of the Korean Society of Computer Information Conference
    • /
    • 한국컴퓨터정보학회 2021년도 제64차 하계학술대회논문집 29권2호
    • /
    • pp.401-403
    • /
    • 2021
  • 본 연구는 한국지능정보사회진흥원에서 제공한 2018년 스마트폰 과의존 실태조사에서 사용된 11개 변수와 스마트폰 과의존과의 관계를 탐색하고, 이를 통해 딥러닝 기반 스마트폰 과의존 분류 분석 모델을 개발하고자 시행되었다. 학습데이터셋은 전국 10,000개 가구내 만 3-69세 스마트폰 이용자 25,465명의 스마트폰 이용 형태 및 개인적 특성에 관한 데이터이다. 딥러닝은 심층신경망(DNN)을 설계하였으며, 은닉층(hidden layer)은 4개층으로 구성하였다. 입력한 데이터는 각각 200개, 150개, 100개, 50개, 2개 노드를 거치면서 최종 출력 정보인 스마트폰 과의존 분류율로 나타나는 모델이다. 이때 스마트폰 과의존 분류률을 높이기 위해 학습률(learning rate)과 같은 하이퍼 파라미터를 활용하여 세부조정하면서 가장 잘 학습하는 값을 찾아내었다. 연구결과, 학습횟수가 300번으로 학습율(learning.rate)이 0.01일때 훈련데이터에서 97.43%, 검증데이터에서 98.06%로 가장 높게 나타났다.

  • PDF

Optimization of Multi-time Scale Loss Function Suitable for DNN-based Audio Coder (심층신경망 기반 오디오 부호화기를 위한 Multi-time Scale 손실함수의 최적화)

  • Shin, Seung-Min;Byun, Joon;Park, Young-Cheol;Beack, Seung-kwon;Sung, Jong-mo
    • Proceedings of the Korean Society of Broadcast Engineers Conference
    • /
    • 한국방송∙미디어공학회 2022년도 하계학술대회
    • /
    • pp.1315-1317
    • /
    • 2022
  • 최근, 심층신경망 기반 오디오 부호화기가 활발히 연구되고 있다. 심층신경망 기반 오디오 부호화기는 기존의 전통적인 오디오 부호화기보다 구조적으로 간단하지만, 네트워크의 복잡도를 증가시키지 않고 인지적 성능향상을 기대하는 것은 어렵다. 이 문제를 해결하기 위하여 인간의 청각적 특성을 활용한 심리음향모델 기반 손실함수를 사용한 기법들이 소개되었다. 심리음향 모델 기반 손실함수를 사용한 오디오 부호화기는 양자화 잡음을 잘 제어하였지만, 여전히 지각적인 향상이 필요하다. 본 논문에서는 심층신경망 기반 오디오 부호화기를 위한 Multi-time Scale 손실함수의 지역 손실함수 윈도우 크기의 최적화 제안한다. Multi-time Scale 손실함수의 지역 손실함수 계산을 위한 윈도우 크기를 조절하며, 이를 통하여 오디오 부호화에 적합한 윈도우 사이즈를 결정한다. 실험을 통해 얻은 최적의 Multi-time Scale 손실함수를 사용하여 네트워크를 훈련하였고, 주관적 평가를 통해 기존의 심리음향모델 기반 손실함수보다 좋은 음성 품질을 보여주는 것을 확인하였다.

  • PDF

Compression of DNN Integer Weight using Video Encoder (비디오 인코더를 통한 딥러닝 모델의 정수 가중치 압축)

  • Kim, Seunghwan;Ryu, Eun-Seok
    • Journal of Broadcast Engineering
    • /
    • 제26권6호
    • /
    • pp.778-789
    • /
    • 2021
  • Recently, various lightweight methods for using Convolutional Neural Network(CNN) models in mobile devices have emerged. Weight quantization, which lowers bit precision of weights, is a lightweight method that enables a model to be used through integer calculation in a mobile environment where GPU acceleration is unable. Weight quantization has already been used in various models as a lightweight method to reduce computational complexity and model size with a small loss of accuracy. Considering the size of memory and computing speed as well as the storage size of the device and the limited network environment, this paper proposes a method of compressing integer weights after quantization using a video codec as a method. To verify the performance of the proposed method, experiments were conducted on VGG16, Resnet50, and Resnet18 models trained with ImageNet and Places365 datasets. As a result, loss of accuracy less than 2% and high compression efficiency were achieved in various models. In addition, as a result of comparison with similar compression methods, it was verified that the compression efficiency was more than doubled.

Speech emotion recognition using attention mechanism-based deep neural networks (주목 메커니즘 기반의 심층신경망을 이용한 음성 감정인식)

  • Ko, Sang-Sun;Cho, Hye-Seung;Kim, Hyoung-Gook
    • The Journal of the Acoustical Society of Korea
    • /
    • 제36권6호
    • /
    • pp.407-412
    • /
    • 2017
  • In this paper, we propose a speech emotion recognition method using a deep neural network based on the attention mechanism. The proposed method consists of a combination of CNN (Convolution Neural Networks), GRU (Gated Recurrent Unit), DNN (Deep Neural Networks) and attention mechanism. The spectrogram of the speech signal contains characteristic patterns according to the emotion. Therefore, we modeled characteristic patterns according to the emotion by applying the tuned Gabor filters as convolutional filter of typical CNN. In addition, we applied the attention mechanism with CNN and FC (Fully-Connected) layer to obtain the attention weight by considering context information of extracted features and used it for emotion recognition. To verify the proposed method, we conducted emotion recognition experiments on six emotions. The experimental results show that the proposed method achieves higher performance in speech emotion recognition than the conventional methods.

DeepBlock: Web-based Deep Learning Education Platform (딥블록: 웹 기반 딥러닝 교육용 플랫폼)

  • Cho, Jinsung;Kim, Geunmo;Go, Hyunmin;Kim, Sungmin;Kim, Jisub;Kim, Bongjae
    • The Journal of the Institute of Internet, Broadcasting and Communication
    • /
    • 제21권3호
    • /
    • pp.43-50
    • /
    • 2021
  • Recently, researches and projects of companies based on artificial intelligence have been actively carried out. Various services and systems are being grafted with artificial intelligence technology. They become more intelligent. Accordingly, interest in deep learning, one of the techniques of artificial intelligence, and people who want to learn it have increased. In order to learn deep learning, deep learning theory with a lot of knowledge such as computer programming and mathematics is required. That is a high barrier to entry to beginners. Therefore, in this study, we designed and implemented a web-based deep learning platform called DeepBlock, which enables beginners to implement basic models of deep learning such as DNN and CNN without considering programming and mathematics. The proposed DeepBlock can be used for the education of students or beginners interested in deep learning.