检索结果-南通市图书馆

Context-Aware Social Media User Sentiment Analysis

在线全文

学校读者我要写书评

暂无评论

Tsinghua Science and Technology 2020年第4期25卷 528-541页

作者： Bo Liu Shijiao Tang Xiangguo Sun Qiaoyun Chen Jiuxin Cao junzhou Luo Shanshan Zhao the School of Computer Science and Engineering Southeast UniversityNanjing 211189China the Key Laboratory of Computer Network and Information of Ministry of Education of China Nanjing 211189China Microsoft Research Asia Suzhou 215000China the School of Cyber Science and Engineering Southeast UniversityNanjing 211189China the Department of Computer Science and Creative Technologies University of the West of EnglandBristolBS161QYUK

The user-generated social media messages usually contain considerable multimodal *** messages are usually short and lack explicit sentiment ***,we can understand the sentiment associated with such messages by analyzing the context,which is essential to improve the sentiment analysis ***,majority of the existing studies consider the impact of contextual information based on a single data *** this study,we propose a novel model for performing context-aware user sentiment *** model involves the semantic correlation of different modalities and the effects of tweet context *** on our experimental results obtained using the Twitter dataset,our approach is observed to outperform the other existing methods in analysing user sentiment.

关键词： social media sentiment analysis multimodal data context-aware topic model

Cooperation in an asymmetric volunteer's dilemma game with relatedness

在线全文

学校读者我要写书评

暂无评论

Chinese Science Bulletin 2012年第16期57卷 1972-1981页

作者： HE junzhou WANG RuiWu CHRISTOPHER X J Jensen LI YaoTang LI ChaoQian School of Mathematics and Statistics Yunnan UniversityKunming 650091China State Key Laboratory of Genetic Resources and Evolution Kunming Institute of ZoologyChinese Academy of SciencesKunming 650223China Department of Mathematics and Science School of Liberal Arts and SciencesPratt Institute200 Willoughby AvenueBrooklynNY 11205USA

What motivates some members of a social group to voluntarily incur costs in order to provide for the common good? This question lies at the heart of theoretical and empirical studies of cooperative behavior. This is also the question that underlies the classic volunteer's dilemma model, which has been previously explored in scenarios where group members are related or interact asym- metrically. Here we present a model that combines asymmetry and relatedness, showing that the probability of volunteerism in such systems depends closely on both the degree of asymmetry and level of relatedness between interacting individuals. As has been shown in previous volunteer's dilemma models, the payoff ratio and overall group size also influence the probability of volunteerism. The probability of volunteerism decreases with increasing group size or decreasing cost-to-benefit ratio of the coplayers; in the presence of asymmetrical interactions, subordinate players were more likely to offer public goods than the dominant player. More asymmetrical interactions decrease the probability of volunteerism of the dominant player; overall volunteerism increases with increasing relatedness.

关键词：不对称性志愿者关联性合作游戏相互作用社会团体关联程度

DeepNoise: Signal and Noise Disentanglement Based on Classifying Fluorescent Microscopy Images via Deep Learning

在线全文

学校读者我要写书评

暂无评论

Genomics, Proteomics & Bioinformatics 2022年第5期20卷 989-1001页

作者： Sen Yang Tao Shen Yuqi Fang Xiyue Wang Jun Zhang Wei Yang junzhou Huang Xiao Han Tencent AI Lab Shenzhen 518057China Department of Electronic Engineering The Chinese University of Hong KongHong Kong Special Administrative Region 999077China College of Computer Science Sichuan UniversityChengdu 610065China

The high-content image-based assay is commonly leveraged for identifying the phenotypic impact of genetic perturbations in biology ***,a persistent issue remains unsolved during experiments:the interferential technical noises caused by systematic errors(e.g.,temperature,reagent concentration,and well location)are always mixed up with the real biological signals,leading to misinterpretation of any conclusion ***,we reported a mean teacher-based deep learning model(Deep Noise)that can disentangle biological signals from the experimental ***,we aimed to classify the phenotypic impact of 1108 different genetic perturbations screened from 125,510 fluorescent microscopy images,which were totally unrecognizable by the human *** validated our model by participating in the Recursion Cellular Image Classification Challenge,and Deep Noise achieved an extremely high classification score(accuracy:99.596%),ranking the 2nd place among 866 participating *** promising result indicates the successful separation of biological and technical factors,which might help decrease the cost of treatment development and expedite the drug discovery *** source code of Deep Noise is available at https://***/Scu-sen/Recursion-Cellular-Image-Classification-Challenge.

关键词： Fluorescent microscopy image Biological signal Classification Deep learning Genetic perturbation

Distributed Storage System for Electric Power Data Based on HBase

在线全文

学校读者我要写书评

暂无评论

Big Data Mining and Analytics 2018年第4期1卷 324-334页

作者： Jiahui Jin Aibo Song Huan Gong Yingying Xue Mingyang Du Fang Dong junzhou Luo School of Computer Science and Engineering Southeast UniversityNanjing 211189China

Managing massive electric power data is a typical big data application because electric power systems generate millions or billions of status,debugging,and error records every single *** guarantee the safety and sustainability of electric power systems,massive electric power data need to be processed and analyzed quickly to make real-time *** solutions typically use relational databases to manage electric power ***,relational databases cannot efficiently process and analyze massive electric power data when the data size increases *** this paper,we show how electric power data can be managed by using HBase,a distributed database maintained by *** system consists of clients,HBase database,status monitors,data migration modules,and data fragmentation *** evaluate the performance of our system through a series of *** also show how HBase’s parameters can be tuned to improve the efficiency of our system.

关键词： electric power data HBase data storage

Effect of Short-term Forest Bathing in Urban Parks on Perceived Anxiety of Young-adults: A Pilot Study in Guiyang, Southwest China

在线全文

学校读者我要写书评

暂无评论

Chinese Geographical Science 2019年第1期29卷 139-150页

作者： ZHOU Changwei YAN Lingbin YU Lifei WEI Hongxu GUAN Haoming SHANG Chongfei CHEN Feiyu BAO junzhou College of Life Science Guizhou University Northeast Institute of Geography and Agricultural Ecology Chinese Academy of Sciences

Forest can be taken as a natural therapy to alleviate perceived anxiety of visitors. Given the geographical difference between urban and rural forest environments, little is known about the urban forest therapy effect of anxiety alleviation with reference to the rural forest. In this study, forty-three university students(aged from 19 to 23) were recruited as participants to visit the forest parks at urban and rural areas of Guiyang City on 21 and 23 December 2016. The forest experience was separated by four sceneries. Participants were asked to complete questionnaires by self-evaluating specific anxiety change from 12 questions with scores from 1 to 10 at both entrance and exit of the parks. The Wilcoxon signed-rank test and paired t-test were used to compare the change of anti-anxiety scores during forest bathing and between urban and rural forests, respectively. Results revealed that forest bathing in the urban park can alleviate the anxiety from financial state(P = 0.0028), exam-pass pressure(P = 0.0040) and love-affair relationship(P = 0.0286). Although rural forest bathing can also alleviate the anxiety from financial state(P = 0.0222), meanwhile, it maintained the anxiety about campus life(P < 0.0001). Forest tree richness tended to be higher in the rural forest park than in the urban one, which in contrast decreased the anxiety alleviation from inter-communication in the rural forest park(P = 0.0487). Principle component analysis indicated that participants tended to perceive more decline of anxiety from social contact in the urban forest. In conclusion, university students were recommended to pay a short visit to the urban forest with partners if they felt anxious about personal affairs and felt necessary to talk with others. For general people's visiting, urban forest trees can be controlled in diversity to some extent to look orderly and alleviate perceived anxiety.

关键词： urbanization tension mental disease behavior disorder undergraduate well-being

Development of upland rice introgression lines and identification of QTLs for basal root thickness under different water regimes

在线全文

学校读者我要写书评

暂无评论

Journal of Genetics and Genomics 2011年第11期38卷 547-556页

作者： junzhou Li Deping Wang Yan Xie Hongliang Zhang Guanglong Hu Jinjie Li Anyong Dai Lifeng Liu Zichao Li Key Laboratory of Crop Heterosis and Utilization （MOE） and Beijing Key Laboratory of Crop Genetic Improvement China Agricultural University Beijing 100193 China College of Agriculture Henan Agricultural University Zhengzhou 450002China China National Centre for Biotechnology Development Beijing 100039 China

Introgression lines （ILs） are valuable materials for identifying quantitative trait loci （QTLs）, evaluating genetic interactions, and marker assisted breeding. A set of 430 ILs （BC5F3） containing segments from upland tropical japonica cultivar IRAT109 in a lowland temperate japonica cultivar Yuefu background were developed. One hundred and seventy-six polymorphic markers were used to identify introgressed segments. No segment from IRAT109 was found in 160 lines. Introgressed segments of the other 270 lines covered 99.1% of the donor genome. The mean number of introgressed donor segments per individual was 3.3 with an average length of 14.4 cM. QTL analysis was conducted on basal root thickness （BRT） of the 270 ILs grown under irrigated lowland, upland and hydroponic conditions. A total of 22 QTLs affecting BRT were identified, six QTLs （qBRT3.1, qBRT3.2, qBRT6.1, qBRTS.2, qBRT9.1, and qBRT9.2） were consistently expressed under at least two environments （location and water regime）, and qBRT7.2 was a new BRT QTL identified under lowland conditions. IL255 containing qBRT9.1 showed an increase of 10.09% and 7.07% BRT over cultivar Yuefu when grown under upland and lowland conditions, respectively. Using a population of 304 F2：3 lines derived from the cross IL255 Yuefu, qBRT9.1 was validated and mapped to a 1.2 cM interval between RM24271 and RM566. The presence of qBRT9.1 explained 12% of BRT variation. The results provide upland rice ILs and BRT QTLs for analyzing the genetic basis of drought resistance, detecting favorable genes from upland rice, and rice drought resistance breeding.

关键词： Upland rice Quantitative trait loci Introgression lines Basal root thickness

An Architectural Modelfor Intelligent Network Management

在线全文

学校读者我要写书评

暂无评论

Journal of Computer Science & Technology 2000年第2期15卷 136-143页

作者：罗军舟顾冠群费翔 Department of Computer Science and Engineering Southeast University Nanjing P.R. China

Traditional network management approach involves the management of each vendor's equipment and network segment in isolation through its own proprietary element management system. It is necessary to set up a new network management architecture that calls for operation consolidation across vendor and technology boundaries. In this paper,an architectural model for Intelligent Network Management (INM) is presented. The INM system includes a manager system, which controls all subsystems and coordinates different management tasks; an expert system, which is responsible for handling particularly difficult problems, and intelligent agents, which bring the management closer to applications and user requirements by spreading intelligent agents through network segments or domain. In the expert system model proposed, especially an intelligent fault management system is given. The architectural model is to build the INM system to meet the need of managing modern network systems.

关键词： network management intelligent network management artificial intelligence expert system intelligent agent

A Study on Genotype-by-Environment Interaction Analysis for Agronomic Traits of Maize Genotypes Across Huang-Huai-Hai Region in China

在线全文

学校读者我要写书评

暂无评论

Phyton-International Journal of Experimental Botany 2022年第1期91卷 57-81页

作者： Haiwang Yue Jianwei Wei Junliang Xie Shuping Chen Haicheng Peng Hongmei Cao junzhou Bu Xuwen Jiang Dryland Farming Institute Hebei Academy of Agriculture and Forestry SciencesHebei Provincial Key Laboratory of Crops Drought Resistance ResearchHengshui053000China Soil and Fertilizer Workstation of Hengshui Agriculture and Rural Bureau Hengshui053000China Maize Research Institute/College of Agronomy Qingdao Agricultural UniversityQingdao266109China

Facing the trend of increasing population,how to increase maize grain yield is a very important issue to ensure food *** this study,28 nationally approved maize hybrids were evaluated across 24 different climatic conditions for two consecutive years(2018-2019).The purpose of this study was to select high-yield with stable genotypes and identify important agronomic traits for maize breeding program *** results of this study showed that the genotypeenvironment interaction effects of the 12 evaluated agronomic traits was highly significant(P<0.001).We introduced a novel multi-trait genotype-ideotype distance index(MGIDI)to select genotypes based on multiple agronomic *** selection process exhibited by this method is unique and easy to understand,so the MGIDI index will have more and more important applications in future multi-environment trials(METs)*** genotypes selected by the MGIDI index were G22,G10,G12 and G1 as the high yielding and stable *** parents of these selected genotypes have the ability to play a greater role as the basic germplasm in the breeding process.A new form of genotype(G)main effects and genotype(G)-by-environment(E)interaction(GGE)technician,genotypeyieldtrait(GYT)biplot,based on multiple traits for genotypes selection was also applied in this *** GYT biplot ranked genotypes by combining grain yield with other evaluated agronomic traits,and displayed the distribution of their traits,namely strengths and weaknesses.

关键词： GEI MGIDI GYT biplot AMMI model GGE

基于主动学习的肺结节计算机辅助诊断交互审查技术

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

生物医学工程学杂志 2024年第3期41卷 503-510页

作者：谭双平李俊张晓娟严馨月张彤吴下里刘自强李莉莉冯娟韩海斌唐国英韩俊洲邓友锋华中科技大学同济医学院附属普爱医院(武汉市第四医院) 武汉430033 武汉大学武汉430079 武汉市江夏区第一人民医院武汉430200

基于电子计算机断层扫描(CT)影像的肺结节自动检测可以有效辅助肺癌诊治,但当前缺乏有效的交互工具将放射科医生的判读结果实时记录并反馈,以优化后台算法模型。本文设计并研发了一个支持CT图像肺结节辅助诊断的在线交互审查系统,通过... 详细信息

基于电子计算机断层扫描(CT)影像的肺结节自动检测可以有效辅助肺癌诊治,但当前缺乏有效的交互工具将放射科医生的判读结果实时记录并反馈,以优化后台算法模型。本文设计并研发了一个支持CT图像肺结节辅助诊断的在线交互审查系统,通过预置模型检测出肺结节展示给医生,医生利用专业知识对检测的肺结节进行标注,然后根据标注结果采用主动学习策略对内置模型进行迭代优化,以持续提高模型的准确性。本文以开源肺结节数据集——肺结节分析2016(LUNA16)的5~9号子集进行迭代实验,随着迭代次数的增加,模型的准确率、调和分数和交并比指标稳定提升,准确率从0.2139提高至0.5656。本文研究结果表明,该系统能在使用深度分割模型辅助医生诊断的同时,最大程度地利用医生的反馈信息来优化模型,迭代提高模型的准确性,从而更好地辅助医生工作。

关键词：肺结节检测主动学习交互审查