检索结果-南通市图书馆

Long Short-Term Memory Recurrent Neural Network-Based Acoustic Model Using Connectionist Temporal Classification on a Large-Scale Training Corpus

引用

China Communications 2017年第9期14卷 23-31页

作者： Donghyun Lee minkyu Lim Hosung park Yoseb Kang Jeong-Sik park Gil-Jin Jang Ji-Hwan Kim Department of Computer Science and Engineering Sogang University Department of Information and Communication Engineering Yeungnam University School of Electronics Engineering Kyungpook National University

A Long Short-Term Memory(LSTM) Recurrent Neural Network(RNN) has driven tremendous improvements on an acoustic model based on Gaussian Mixture Model(GMM). However, these models based on a hybrid method require a forced aligned Hidden Markov Model(HMM) state sequence obtained from the GMM-based acoustic model. Therefore, it requires a long computation time for training both the GMM-based acoustic model and a deep learning-based acoustic model. In order to solve this problem, an acoustic model using CTC algorithm is proposed. CTC algorithm does not require the GMM-based acoustic model because it does not use the forced aligned HMM state sequence. However, previous works on a LSTM RNN-based acoustic model using CTC used a small-scale training corpus. In this paper, the LSTM RNN-based acoustic model using CTC is trained on a large-scale training corpus and its performance is evaluated. The implemented acoustic model has a performance of 6.18% and 15.01% in terms of Word Error Rate(WER) for clean speech and noisy speech, respectively. This is similar to a performance of the acoustic model based on the hybrid method.

关键词： acoustic model connectionisttemporal classification large-scale trainingcorpus long short-term memory recurrentneural network

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Changes in spatial variations of sap flow in Korean pine trees due to environmental factors and their effects on estimates of stand transpiration

引用

Journal of Mountain Science 2016年第6期13卷 1024-1034页

作者： minkyu MOON Taek-yu KIM Juhan park Sungsik CHO Daun RYU Sanguk SUH Hyun Seok KIM National Center for Agro Meteorology 1 Gwanak-roGwanak-guSeou1151-744Republic of Korea National Institute of Environmental Research 42 Hwangyong-roSeo-guIncheon 404-780Republic of Korea Department of Forest Sciences Seoul National UniversityI Gwanak-roGwanak-guSeou1151-921Republic of Korea National Academy of Agricultural Science 166 Nongsaengmyeong-roWanju-gun 565-851Republic of Korea Research Institute for Agriculture and Life Sciences 1 Gwanak-roGwanak-guSeou1151-921Republic of Korea

It is difficult to scale up measurements of the sap flux density(J_S) for the characterization of tree or stand transpiration(E) due to spatial variations in J_S and their temporal *** assess spatial variations in the sap flux density of Korean pine(Pinus koraiensis) and their effects on E estimates,we measured the J_S using Granier-type *** trees,the J_S decreased exponentially with the radial depth,and the J_S of the east aspects were higher than those of the west *** trees,there was a positive relationship between J_S and the tree diameter at breast height,and this positive relationship became stronger as the transpiration demand *** spatial variations that caused large errors in E estimates(i.e.,up to 110.8 % when radial variation was ignored) had varied systematically with environmental factors systematic characteristics in relation to environmental ***,changes in these variations did not generate substantial errors in the E *** our study periods,the differences in the daily E(E_D) calculated by ignoring radial,azimuthal and tree-to-tree variations and the measured E_D were fairly constant,especially when the daily vapor pressure deficit(D_D)was higher than 0.6 k *** results imply that the effect of spatial variations changes on sap flow can be a minor source of error compared with spatial variations(radial,azimuthal and tree-to-tree variations) when considering E estimates.

关键词：空间变化树干液流蒸腾量环境因子林分径向变异松树韩国

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Deep-learning-based gas identification by timevariant illumination of a single micro-LEDembedded gas sensor

引用

Light(Science & Applications) 2023年第5期12卷 826-837页

作者： Incheol Cho Kichul Lee Young Chul Sim Jae-Seok Jeong minkyu Cho Heechan Jung Mingu Kang Yong-Hoon Cho Seung Chul Ha Kuk-Jin Yoon Inkyu park Department of Mechanical Engineering Korea Advanced Institute of Science and Technology(KAIST)291 Daehak-roYuseong-guDaejeon 34141Republic of Korea Department of Physics Korea Advanced Institute of Science and Technology(KAIST)291 Daehak-roYuseong-guDaejeon 34141Republic of Korea SENKO Co. Ltd.485Oesammi-DongOsan-SiGyeonggil-Do 18111Republic of Korea

Electronic nose(e-nose)technology for selectively identifying a target gas through chemoresistive sensors has gained much attention for various applications,such as smart factory and personal health *** overcome the crossreactivity problem of chemoresistive sensors to various gas species,herein,we propose a novel sensing strategy based on a single micro-LED(μLED)-embedded photoactivated(μLP)gas sensor,utilizing the time-variant illumination for identifying the species and concentrations of various target gases.A fast-changing pseudorandom voltage input is applied to the μLED to generate forced transient sensor responses.A deep neural network is employed to analyze the obtained complex transient signals for gas detection and concentration *** proposed sensor system achieves high classification(~96.99%)and quantification(mean absolute percentage error~31.99%)accuracies for various toxic gases(methanol,ethanol,acetone,and nitrogen dioxide)with a single gas sensor consuming 0.53 *** proposed method may significantly improve the efficiency of e-nose technology in terms of cost,space,and power consumption.

关键词： Deep illumination transient

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Preoperative serum carbohydrate antigen 19-9 levels predict early recurrence after the resection of early-stage pancreatic ductal adenocarcinoma

引用

World Journal of Gastrointestinal Surgery 2021年第11期13卷 1423-1435页

作者： Sarang Hong Ki Byung Song Dae Wook Hwang Jae Hoon Lee Woohyung Lee Eunsung Jun Jaewoo Kwon Yejong park Seo Young park Naru Kim Dakyum Shin Hyeyeon Kim minkyu Sung Yunbeom Ryu Song Cheol Kim Division of Hepatobiliary and Pancreatic Surgery Department of SurgeryAsan Medical CenterSeoul 05505South Korea Department of Surgery Kangbuk Samsung HospitalSungkyunkwan University School of MedicineSeoul 03181South Korea Department of Statistics and Data Science Korea National Open UniversitySeoul 03087South Korea Department of Surgery Uijeongbu St.Mary's HospitalCollege of MedicineGyeonggido 11765South Korea

BACKGROUND Pancreatic ductal adenocarcinoma(PDAC)is a serious disease with a poor *** a minority of patients undergo surgery due to the advanced stage of the disease,and patients with early-stage disease,who are expected to have a better prognosis,often experience ***,it is important to identify the risk factors for early recurrence and to develop an adequate treatment *** To evaluate the predictive factors associated with the early recurrence of earlystage *** This study enrolled 407 patients with stage I PDAC undergoing upfront surgical resection between January 2000 and April *** recurrence was defined as a diagnosis of recurrence within 6 mo of *** optimal cutoff values were determined by receiver operating characteristic(ROC)*** and multivariate analyses were performed to identify the risk factors for early *** Of the 407 patients,98 patients(24.1%)experienced early disease recurrence:26(26.5%)local and 72(73.5%)distant *** total,253(62.2%)patients received adjuvant *** ROC curve analysis,the optimal cutoff values for early recurrence were 70 U/mL and 2.85 cm for carbohydrate antigen 19-9(CA 19-9)levels and tumor size,*** the 181 patients with CA 19-9 level>70 U/mL,59(32.6%)had early recurrence,compared to 39(17.4%)of 226 patients with CA 19-9 level≤70 U/mL(P70 U/mL(P=0.006),tumor size>2.85 cm(P=0.004),poor differentiation(P=0.008),and non-adjuvant chemotherapy(P=0.025)were significant risk factors for early recurrence in early-stage *** Elevated CA 19-9 level(cutoff value>70 U/mL)can be a reliable predictive factor for early recurrence in early-stage *** adjuvant chemotherapy can prevent early recurrence,it should be recommended for patients susceptible to early recurrence.

关键词： Pancreatic ductal adenocarcinoma Early recurrence Upfront surgery Carbohydrate antigen 19-9 Adjuvant chemotherapy

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

A MODIFIED FULL-ARRAY ARCHITECTURE FOR 64 x 64 BIT MULTIPLIER

A MODIFIED FULL-ARRAY ARCHITECTURE FOR 64 x 64 BIT MULTIPLIE...

引用

The Third International Conference on Solid State and Integrated Circuit Technology

作者： minkyu Song Hocheol park Changwoo Jee Jongsoo Lee G4FAX Lab. KAITECH SEOUL KOREA G4FAX Lab. KAITECH SEOUL KOREA G4FAX Lab. KAITECH SEOUL KOREA G4FAX Lab. KAITECH SEOUL KOREA

＜正＞ In this paper,a 64 x 64 bit modified full-array multiplier using a new compression technique is *** multiplier is based on a modular array architecture with the block of an encoder to implement the modified Booth’s algorithm [1],the block implementing the Wallace’s tree [2],and the block of a 128-bit Carry Look-Ahead（CLA） *** input of an encoder is 16-bit long and the block implementing the Wallace’s tree consists of both 4-2 compressors and 6-2 compressors proposed in this *** those compressors,the propagation delay time of the proposed multiplier is about 5% reduced,compared to the multiplier which has a parallel structured architecture [4].Also the multiplier has easier design methodology due to the regular characteristics of the modular array architecture.

关键词：

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

基于辣椒基因组测序探究辣味在种群中的进化

引用

辣椒杂志 2014年第2期12卷 40-50页

作者： Seungill Kim minkyu park Seon-In Yeom Yong-Min Kim Je Min Lee Hyun-Ah Lee Eunyoung Seo Jaeyoung Choi Kyeongchae Cheong Ki-Tae Kim Kyongyong Jung Gir-Won Lee Sang-Keun Oh Chungyun Bae Saet-Byul Kim Hye-Young Lee Shin-Young Kim Myung-Shin Kim Byoung-Cheorl Kang Yeong Deuk Jo Hee-Bum Yang Hee-Jin Jeong Won-Hee Kang Jin-Kyung Kwon Chanseok Shin Jae Yun Lim June Hyun park Jin Hoe Huh June-Sik Kim Byung-Dong Kim Oded Cohen Ilan Paran Mi Chung Suh Saet Buyl Lee Yeon-Ki Kim Younhee Shin Seung-Jae Noh Junhyung park Young Sam Seo Suk-Yoon Kwon Hyun A Kim Jeong Mee park Hyun-Jin Kim Sang-Bong Choi Paul W Bosland Gregory Reeves Sung-Hwan Jo Bong-Woo Lee Hyung-Taeg Cho Hee-Seung Choi Min-Soo Lee Yeisoo Yu Yang Do Choi Beom-Seok park Allen van Deynze Hamid Ashrafi Theresa Hill Woo Taek Kim Hyun-Sook Pai Hee Kyung Ahn Inhwa Yeam James J Giovannoni Jocelyn K C Rose Iben Sφrensen Sang-Jik Lee Ryan W Kim Ik-Young Choi Beom-Soon Choi Jong-Sung Lim Yong-Hwan Lee Doil Choi 刘峰张西露 Department of Plant Science Seoul National University Plant Genomics and Breeding Institute Seoul National University Department of Agricultural Biotechnology Seoul National University Department of Bioinformatics and Life Science Soongsil University Vegetable Breeding Research Center Seoul National University Agricultural Research Organization Institute of Plant Science Volcani Center Bet DaganIsrael. Agricultural Research Organization Institute of Plant Science Volcani Center Bet DaganIsrael Department of Bioenergy Science and Technology Chonnam National University Genomics Genetics Institute GreenGene BioTech Inc. Codes Division Insilicogen Inc. Ginseng Resources Research Laboratory Korea Ginseng Corporation Korea Research Institute of Bioscience and Biotechnology Division of Bioscience and Bioinformatics Myongji University Department of Plant and Environmental Sciences New Mexico State University Chile Pepper Institute New Mexico State University Seeders Inc. Department of Biological Sciences Seoul National University Arizona Genomics Institute University of Arizona Agricultural Genome Center National Academy of Agricultural Science Rural Development Administration Seed Biotechnology Center University of California Davis Davis California USA Department of Systems Biology Yonsei University Department of Horticulture and Breeding Andong National University US Department of Agriculture-Agricultural Research Service Robert W. Holley Center Ithaca New York USA Boyce Thompson Institute for Plant Research Cornell University Ithaca New York USA Department of Plant Biology Cornell University Ithaca New York USA Biotechnology Institute Nongwoo Bio Yeoju Korea Genome Center University of California Davis Davis California USA National Instrumentation Center for Environmental Management Seoul National University Seoul Korea 湖南省蔬菜研究所

辣椒(Capsicum annuum)是美洲最古老的驯化作物之一,在全世界被广泛种植。本研究对CM334(墨西哥地方品种)进行了全基因组测序和组装(覆盖度为186.6×),同时还对另外两个栽培种和一个野生种分别进行了重测序和从头(de novo)测序。结果表... 详细信息

辣椒(Capsicum annuum)是美洲最古老的驯化作物之一,在全世界被广泛种植。本研究对CM334(墨西哥地方品种)进行了全基因组测序和组装(覆盖度为186.6×),同时还对另外两个栽培种和一个野生种分别进行了重测序和从头(de novo)测序。结果表明辣椒基因组大小约为番茄的4倍,辣椒基因组上存在大量gypsy转座子和花椰菜病毒家族元件的积累。结合转录组综合分析认为,辣椒素合成酶基因表达模式的改变及新功能化是导致辣椒素生物合成的原因,研究还发现辣椒与番茄在乙烯合成及果实成熟调控方面存在明显不同的分子模式。辣椒参考基因组的发布为辣椒营养品质和药用价值的改善提供了重要平台。

关键词：基因组测序辣椒素参考基因 Capsicum 转座子辣椒果实分子模式生物合成基因表达模式基因组大小

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

欢迎您,

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：