Understand Students Feedback Using Bi-Integrated CRF Model Based Target Extraction

Educational institutions showing interest to find the opinion of the students about their course and the instructors to enhance the teaching-learning process. For this, most research uses sentiment analysis to track students’ behavior. Traditional sentence-level sentiment analysis focuses on the whole sentence sentiment. Previous studies show that the sentiments alone are not enough to observe the feeling of the students because different words express different sentiments in a sentence. There is a need to extract the targets in a given sentence which helps to find the sentiment towards those targets. Target extraction is the subtask of targeted sentiment analysis. In this paper, we proposed the innovative model to find the targets of the given sentence using Bi-Integrated Conditional Random Fields (CRF). A Parallel fusion neural network model is designed to perform this task. We evaluate the model using the Michigan dataset and we build a dataset for target extraction from student reviews. The experimental results show that our proposed fusion model achieves better results compared to baseline models.     

基于词性特征的明喻识别及要素抽取方法

比喻是一种利用事物之间的相似点建立关系的修辞方式。明喻是比喻中最常见的形式,具有明显的喻词,例如"像",用于关联本体和喻体。近年来高考语文散文类鉴赏题中多有考查明喻句的试题,为了解答此类鉴赏题,需要识别比喻句中的本体和喻体要素。该文提出了基于词性特征的明喻识别及要素抽取方法。首先将句子中词向量化表示与词性特征向量化表示进行融合,将融合后的向量输入到BiLSTM中进行训练,然后利用CRF解码出全局最优标注序列;最后得到明喻识别和要素抽取的结果。公开数据集上的实验结果表明,该方法优于已有的单任务方法;同时也将该文方法应用于北京高考语文鉴赏题中比喻句的识别与要素抽取,验证了方法的可行性。     

Carbonaceous microsphere-based superabsorbent polymer as filler for coating of NPK fertilizer: Fabrication,properties, swelling,and nitrogen release characteristics

Recently, the use of controlled release fertilizers in agriculture has resulted in huge benefits in plant growth and cultivation. Superabsorbent polymer (SAP)-coated fertilizers have the added advantage in retaining water in soil after irrigation and also reduce the nutrient release rate from soil in a controlled manner. This study aimed to produce a nitrogen–phosphorus–potassium (NPK) fertilizer coated with superabsorbent carbonaceous microspheres polymer (SPC) by inverse suspension polymerization method with water-retention and controlled release properties. Two sets of experiments were conducted: (1) three different weight percentages and (2) different materials. NPK coated with SPC showed increasing water-retention ability with respect to carbon microsphere percentages and retains >80% water at the 30th day of experiment compared with pure NPK and NPK coated with SAP. The slow release behavior of all samples was investigated by induced coupled plasma mass spectrometry spectrometry and results showed that NPK coated with SAP and SPC has a low release rate with <50% nutrient release compared with uncoated NPK at the 30th day. The release mechanism kinetics of NPK coated with SAP and SPC were studied based on the Kosmeyer–Peppas model. The mechanisms approached Fickian diffusion-controlled release as the n value for both samples was less than 0.5. © 2019 Wiley Periodicals, Inc. J. Appl. Polym. Sci. 2020 , 137, 48396.     

平面交叉口交通控制与管理不安全因素分析

在现场调查分析交通标志、标线、信号设置存在问题,视距不良、交叉路口交通组织和渠化不当等的基础上,总结了公路平面交叉口在交通控制与管理方面常见的9个不安全因素。运用事故折减系数(CRF)统计数据分析了不安全因素引起事故的可能性,采用汽车碰撞理论分析与专家调查结合的方法,进行各事故类型的严重性分析,进一步确定了各不安全因素的危险性指数,得到其危险性排序,为交叉口安全诊断和安全保障研究提供了依据。     

临床试验逻辑核查的分类及应用

临床试验数据的清理是临床试验过程中非常重要的环节,完整、清洁的数据才能用于统计分析。本文旨在讨论临床试验数据清理过程中数据核查计划的撰写及逻辑核查的分类等,并进一步讨论近年来临床试验数字化趋势对数据的采集及清理可能带来的影响。     

基于多级规约CRF 模型景点评论观点分析方法

针对旅游发展趋势及旅游景点评论观点抽取的难题,提出一种基于机器学习的多级规约条件随机场 (conditional random field,CRF)模型的旅游景点评论观点分析方法。分析常见观点分析方法,利用CRF 算法从数据 集中提取评价观点,基于评论者关系特征构建景点虚假评论者自动识别模型对其评论观点进行过滤,并对评论观点 的相关性进行观点合并。实验结果表明：该方法较其他常见方法在准确率、召回率方面有明显提高,且查准率-召回 率曲线(precision and recall,PR)及接受者操作特征(receiver operating characteristic,ROC)曲线效果良好,对游客选 择旅游目标具有很好的决策作用。     

Enhancing TRU burning and Am transmutation in Advanced Recycling Reactor

This paper presents about conceptual designs of Advanced Recycling Reactor (ARR) focusing on enhancement in transuranics (TRU) burning and americium (Am) transmutation. The design has been conducted in the context of the Global Nuclear Energy Partnership (GNEP) seeking to close nuclear fuel cycle in ways that reduce proliferation risks, reduce the nuclear waste in the US and further improve global energy security. This study strives to enhance the TRU burning and the Am transmutation, assuming the development of related technologies in this study, while the ARR based on mature technologies was designed in the previous study. It has followed that the provided TRU burning core is designed to burn TRU at 28 kg/TW_thh, by adding moderator pins of B₄C (Enriched B-11) and the Am transmutation core will be able to transmute Am at 34 kg/TW_thh, by locating Am blanket of AmN around the TRU burning core. It indicates that these concepts improve TRU burning by 40-50% than the previous core and can transmute Am effectively, keeping the void reactivity acceptable.     

基于完全联系的条件随机场的图像标注

传统的图像标注模型通常存在两个问题：只能够对短距离的像素上下文信息进行建模和复杂的模型推理过程。为了提高图像标注的精度、简化图像标注的模型推理过程,采用完全联系的条件随机场模型进行图像标注,提出利用基于高斯kd树的平均场估计方法实现该模型的高效推理。为了更好地验证算法的有效性,实验的图片数据库不仅包含标准的图片库--剑桥大学微软研究图片库（MSRC-9）,还包含作者制作的机械零件图片库（MyDataset_1）和办公桌图片库（MyDataset_2）。新算法在三个图片库上的平均标注精度分别可以达到77.96%、97.15%和95.35%,每幅图的平均运行时间为2s。实验结果表明,基于完全联系的条件随机场的图像标注能够更充分地考虑不同的像素上下文信息来提高标注精度,而基于高斯kd树的模型推理能够提高模型推理的效率。     

融合规则与统计的微博新词发现方法

结合微博新词的构词规则自由度大和极其复杂的特点,针对传统的C/NC-value方法抽取的结果新词边界的识别准确率不高,以及低频微博新词无法正确识别的问题,提出了一种融合人工启发式规则、C/NC-value改进算法和条件随机场（CRF）模型的微博新词抽取方法。一方面,人工启发式规则是指对微博新词的分类和归纳总结,并从微博新词构词的词性（POS）、字符类别和表意符号等角度设计的微博新词的构词规则;另一方面,改进的C/NC-value方法通过引入词频、邻接熵和互信息等统计量来重构NC-value目标函数,并使用CRF模型训练和识别新词,最终达到提高新词边界识别准确率和低频新词识别精度的目的。实验结果显示,与传统方法相比,所提出的方法能有效地提高微博新词识别的F值。     

面向产品评论的细粒度情感分析

针对传统粗粒度情感分析忽略具体评价对象,以及现有细粒度情感分析方法忽略无关评价要素的问题,提出结合条件随机场(CRF)和语法树剪枝的方法对产品评论进行细粒度情感分析。采用基于MapReduce的并行化协同训练(Tri-training)的方法对语料进行半自主标注,利用融合多种语言特征的条件随机场模型,获取评论中的评价对象和正负面评价词。通过建立领域本体和句法路径库实现语法树剪枝,对含有多个评价对象和评价词的文本,去掉无关评价对象的干扰,抽取出正确的评价单元,最后形成可视化产品报告。实验结果显示,提出的方法在两种不同领域数据集上,识别情感要素的综合准确率达89%左右,情感评价单元的综合准确率也达89%左右。实验结果表明,与传统方法相比,结合CRF和语法树剪枝的方法识别准确率更高,性能更好。