基于词向量模型的情感分析

随着移动互联网的发展,以商品评论等带有主观性的短文本信息急剧增加.海量的文本信息使得人工管理越来越困难.本文以商品评论为研究对象进行情感分析.针对商品评论为短文本的特点,本文在词向量的基础上提出了词向量叠加方法和加权词向量方法进行文本特征的提取,从而更深层次的提取短文本特征.在进行评论情感分析模型性能的比较中,说明了本文所提方法的有效性.基于情感分析技术可以解决人工难以胜任的海量商品评论的分类,方便用户快速获取有效信息.     

JIT在汽车再制造工程中的应用研究

提出了汽车再制造工程的内涵及其社会经济意义,对汽车再制造工程中的关键问题及解决方案进行了分析。针对汽车再制造工程中的实际问题,对利用JIT技术来实现汽车再制造的准时化生产进行了进一步的研究。     

浅谈框架填充墙裂缝的成因及防治措施

结合实际情况,归纳了填充墙常见裂缝的位置和特征,分析了各类型裂缝形成的机理,并对材料、设计和施工提出了具体的改进意见,制定了相关措施,以使裂缝得到有效控制。     

试论企业与高校构建战略联盟的思考

战略联盟是指两个或两个以上具有资源优势互补的企业(特定事业和职能部门)之间,出于对整个市场的预期和企业自身总体经营目标、经营风险的考虑,为达到共同拥有市场、共同使用资源等战略目标通过协议而结成的优势互补、风险共担、利益共享、要素双向或多向流动的松散型网络组织.分析现阶段高校中硕士教育的特点以及企业在技术开发时所面临的问题,提出企业和高校构建战略联盟的思路.分析实施战略联盟的优势和所存在的问题,使企业和高校能够扬长避短,充分的发挥优势合作,注重问题的解决,从而形成一种双赢的局面.     

The Impact of Silicon on Photosynthetic and Biochemical Responses of Sugarcane under Different Soil Moisture Levels

Silicon - Increasing drought stress is one of the most limiting factors for agricultural crop productivity across the world. Silicon (Si) has been known to augment plant protection against drought...     

基于LSTM网络的大雾临近预报模型及应用

长短期记忆网络（LSTM）是一种时间递归神经网络,适合于预测时间序列延续性相对较长的事件.本文基于LSTM网络构建了一个全新的大雾临近预报框架,首先将地面气象要素观测资料转化成时间序列数据,并基于此序列进行建模.为了验证提出的模型的准确性,将安徽省81个国家站近2年地面气象要素数据转换为序列数据,基于该数据集对未来1–4小时进行逐小时大雾预报实验,实验结果显示本文提出的模型其TS-Score分别为61%、55%、36%和31%,明显优于卷积神经网络（CNN）以及传统机器学习算法如支持向量机（SVM）和K-近邻算法（KNN）的预测结果,是大雾临近预报的一种有效预报方法.     

基于Twitter的流感疫情可视化系统

介绍了当下社交媒体的蓬勃发展以及基于社交媒体数据的多领域应用,阐述了利用Twitter数据对流感疫情监测的优势和可行性,着重论述了基于Twitter的流感疫情可视化系统的详细设计和具体实现,并讨论了系统的意义和应用方向.     

Augmented reality and competition in robotics education: Effects on 21st century competencies,group collaboration and learning motivation

Robotics education has received an increasing attention in recent years as a means to build students' motivation, team collaboration skills, and other valuable 21st century competencies. Yet there is a lack of experimental studies to investigate and identify strategies to facilitate robotics education. This study adopted a 2 × 2 quasi-experimental design to investigate two strategies: the incorporation of augmented reality (AR) and the introduction of competition in robotics activities. Students' robotics task performance, team collaboration processes, 21st century learning competencies and learning motivation were measured as dependent variables. The results indicated that AR significantly improved students' motivation, team processes, and 21st century competencies. Moreover, the effects of AR were more pronounced with the competition groups. Implications are drawn to provide guidelines on the use of AR and competition in robotics education.     

基于迁移学习的信用评分预测

在互联网金融机构有很多信贷业务, 部分新开展的业务由于客户数据较少, 无法建立有效的信用评分模型. 本文研究将迁移学习思想应用到该问题中, 利用已有其他业务的客户数据帮助新开展的业务建立有效的信用评分模型. 本文提出一种联合Triplet-Loss表征学习和领域适配的深度学习方法对已有业务数据进行重新编码, 并将重新编码后所得的知识迁移到新开展业务的模型中, 最后使用XGBoost做为分类器. 针对上述问题, 本文提出的模型相对传统机器学习方法在效果上有一定提升, 在一定程度上解决了该问题.     

Development and application of online Stelmor Controlled Cooling System

An online Stelmor Controlled Cooling System (SCCS) has been developed successfully for the Stelmor production line, which can communicate with the material flow management system and Program Logic Control System (PLCs) automatically through local network. This online model adopts Implicit Finite Difference Time Domain (FDTD) method to calculate temperature evolution and phase transformation during the production process and predicts final properties. As Continuous Cooling Temperature (CCT) curves of various steels can be coupled in the model, it can predict the latent heat rise and range of phase transformation for various steels, which can provide direct guidance for new steel development and optimization of present Stelmor cooling process. This unique online system has been installed in three Stelmor production lines at present with good results.