A hybrid knowledge and ensemble classification approach for prediction of venous thromboembolism

Clinical narratives such as progress summaries, lab reports, surgical reports, and other narrative texts contain key biomarkers about a patient's health. Evidence-based preventive medicine needs accurate semantic and sentiment analysis to extract and classify medical features as the input to appropriate machine learning classifiers. However, the traditional approach of using single classifiers is limited by the need for dimensionality reduction techniques, statistical feature correlation, a faster learning rate, and the lack of consideration of the semantic relations among features. Hence, extracting semantic and sentiment-based features from clinical text and combining multiple classifiers to create an ensemble intelligent system overcomes many limitations and provides a more robust prediction outcome. The selection of an appropriate approach and its interparameter dependency becomes key for the success of the ensemble method. This paper proposes a hybrid knowledge and ensemble learning framework for prediction of venous thromboembolism (VTE) diagnosis consisting of the following components: a VTE ontology, semantic extraction and sentiment assessment of risk factor framework, and an ensemble classifier. Therefore, a component-based analysis approach was adopted for evaluation using a data set of 250 clinical narratives where knowledge and ensemble achieved the following results with and without semantic extraction and sentiment assessment of risk factor, respectively: a precision of 81.8% and 62.9%, a recall of 81.8% and 57.6%, an F measure of 81.8% and 53.8%, and a receiving operating characteristic of 80.1% and 58.5% in identifying cases of VTE.     

基于微博用户兴趣话题的相似用户挖掘

相似用户挖掘是提高社交网络服务质量的重要途径，在面向大数据的社交网络时代，准确的相似用户挖掘对于用户和互联网企业等都有重要的意义，而根据用户自己的兴趣话题挖掘的相似用户更符合相似用户的要求。提出了一种基于用户兴趣话题进行相似用户挖掘的方法。该方法首先使用TextRank话题提取方法对用户进行兴趣话题提取，再对用户发表内容进行训练，计算出所有词之间的相似度。提出CP（Corresponding Position similarity）、CPW（Corresponding Position Weighted similarity）、AP（All Position similarity）、APW（All Position Weighted similarity）四种用户兴趣话题词相似度计算方法，通过用户和相似用户间关注、粉丝重合率验证相似用户挖掘效果，APW similarity的相似用户的关注/粉丝重合百分比为1.687%，优于提出的其他三种算法，分别提高了26.3%、2.8%、12.4%，并且比传统的文本相似度方法Jaccard相似度、编辑距离算法、余弦相似度分别提高了20.4%、21.2%、45.0%。因此APW方法可以更加有效地挖掘出用户的相似用户。     

基于查询的新闻多文档自动摘要技术研究

针对新闻文本领域,该文提出一种基于查询的自动文本摘要技术,更加有针对性地满足用户信息需求。根据句子的TF-IDF、与查询句的相似度等要素,计算句子权重,并根据句子指示的时间给定不同的时序权重系数,使得最近发生的新闻内容具有更高的权重,最后使用最大边界相关的方法选择摘要句。通过与基于TF-IDF、Text-Rank、LDA等六种方法的对比,该摘要方法ROUGE评测指标上优于其他方法。从结合评测结果及摘要示例可以看出,该文提出的方法可以有效地从新闻文档集中摘取核心信息,满足用户查询内容的信息需求。     

Evolution of Local Structural Ordering and Chemical Distribution upon Delithiation of a Rock Salt–Structured Li1.3Ta0.3Mn0.4O2 Cathode

Lithium‐rich disordered rock‐salt oxides have attracted great interest owing to their promising performance as Li‐ion battery cathodes. While experimental and theoretical efforts are critical in advancing this class of materials, a fundamental understanding of key property changes upon Li extraction is largely missing. In the present study, single‐crystal synthesis of a new disordered rock‐salt cathode material, Li_1.3Ta_0.3Mn_0.4O₂ (LTMO), and its use as a model compound to investigate Li concentration–driven evolution of local cationic ordering, charge compensation, and chemical distribution are reported. Through the combined use of 2D and 3D X‐ray nanotomography, it is shown that Li removal accompanied by oxygen oxidation is correlated with the development of morphological defects such as particle cracking. Chemical heterogeneity, quantified by subparticle level distribution of Mn valence state, is minimal during Mn redox, which drastically increases upon the formation of cracks during oxygen redox. Density functional theory and bond valence sum mismatch calculations reveal the presence of local short‐range ordering in the pristine oxide, which gradually disappears along with the extraction of Li. The study suggests that with cycling the transformation into true cation–disordered state can be expected, which likely impacts the voltage profile and obtainable energy density of the oxide cathodes.     

Investigation on external stimuli engendered magnetic ordering in polycrystalline CaCu3Ti4O12 quadruple perovskite

The consequences of high energy mechanical milling, microwave-assisted heating and rapid thermal cooling on magnetic ordering in polycrystalline CaCu₃Ti₄O₁₂ cubic perovskite have been investigated by means of X-ray powder diffractometry (300?K), dc magnetization in field – cooled and zero – field cooled modes (H = 100?Oe and 1000?Oe, T?=?5–300?K) (M – T curves) and M – H loop characteristics (T?=?5?K and 300?K, H_max = 70?kOe). The M – T curves of unmilled and 16?h milled samples show pure antiferromagnetic and weak ferromagnetic ordering, respectively, 1?h and 6?h milled samples demonstrate the coexistence of both the phases while microwave-assisted and quenched samples exhibit classic antiferromagnetic transition and a low temperature paramagnetic–like contribution with different weights, well supported by the M – H loop characteristics. The observed transformations in the magnetic ordering are attributed to the ball-milling induced stress which curtails hybridization of empty Ti-3d orbitals with Cu-3d and O-2p orbitals and secondary phase formation. Oxygen vacancies associated with bound magnetic polarons originate ferromagnetism in the milled samples while unpaired electrons inhabited at the empty sites are the cause of paramagnetic centers. The low-temperature Curie – tail in M – T curve for quenched and microwave assisted samples is attributed to Ti³⁺ cations.     

基于门控联合池化自编码器的通用性文本表征

为了学习文本的语义表征,以往的研究者主要依赖于复杂的循环神经网络(recurrent neural networks, RNNs)和监督式学习方法。该文提出了一种门控联合池化自编码器(gated mean-max AAE)用于学习中英文的文本语义表征。该文的自编码器完全通过多头自注意力机制(multi-head self-attention mechanism)来构建编码器和解码器网络。在编码阶段,提出了均值—最大化(mean-max)联合表征策略,即同时运用平均池化(mean pooling)和最大池化(max pooling)操作来捕获输入文本中多样性的语义信息。为促使联合池化表征可以全面地指导重构过程,解码器采用门控操作进行动态关注。通过在大规模中英文未标注语料上训练模型,获得了高质量的句子编码器。在重构文本段落的实验中,该文模型在实验效果和计算效率上均超越了传统的RNNs模型。将公开训练好的文本编码器,使其可以方便地运用于后续的研究。     

基于重量分析的OBDD变量排序算法

有序的二叉判决图（ＯＢＤＤ）是布尔表达式的一种有效表示方法，但它的体积对变量排序具有较强的依赖性。本文提出一种电路结构图，并在此基础上定义了原始输入重量和节点重量等参数，并建立了用重量分析来指导的ＯＢＤＤ变量排序算法。由于从考虑变量对输出函数的影响出发与从考虑ＯＢＤＤ节点共享性出发对变量排序的要求不同，本文分别设计了两类算法。实验结果表明，本文对大多数标准电路变量排序的效果都优于国际上的同类算法，     

Industrial Text and French Neo-structuralism

Parallel to, and to some degree inreaction to French poststructuralisttheorization (as championed by Derrida,Foucault, and Lacan, among others) is a Frenchneo-structuralism built directly on theachievements of structuralism using electronicmeans. This paper examines some exemplaryapproaches to text analysis in thisneo-structuralist vein: SATOR's topoidictionary, the WinBrill POS tagger andFrançois Rastier's interpretativesemantics. I consider how a computer-assisted``Wissenschaft' accumulation of expertisecomplements the neo-structuralist approach.Ultimately, electronic critical studies will bedefined by their strategic position at theintersection of the two chief technologiesshaping our society: the new informationprocessing technology of computers and therepresentational techniques that haveaccumulated for centuries in texts.Understanding how these two informationmanagement paradigms complement each other is akey issue for the humanities, for computerscience, and vital to industry, even beyond thenarrow realm of the language industries. Thedirection of critical studies, a small planetlong orbiting in only rarefied academiccircles, will be radically altered by the sheersize of the economic stakes implied by a newkind of text, the industrial text, thetechnological heart of an information society.     

Fuzzy Weirstrass theorem and convex fuzzy mappings

The convexity and continuity of fuzzy mappings are defined through a linear ordering and a metric on the set of fuzzy numbers. The local-global minimum property of real-valued convex functions is extended to convex fuzzy mappings. It is proved that a strict local minimizer of a quasiconvex fuzzy mapping is also a strict global minimizer. Characterizations for convex fuzzy mappings and quasiconvex fuzzy mappings are given. In addition, the Weirstrass theorem is extended from real-valued functions to fuzzy mappings.     

Wavefront reduction using graphs,neural networks and genetic algorithm

This paper presents an algorithm for nodal numbering in order to obtain a small wavefront. Element clique graphs are employed as the mathematical models of finite element meshes. A priority function containing five vectors is used, which can be viewed as a generalization of Sloan's function. These vectors represent different connectivity properties of the graph models. Unlike Sloan's algorithm, which uses two fixed coefficients, here, five coefficients are employed, based on an evaluation by artificial neural networks. The networks weights are obtained using a simple genetic algorithm. Examples are included to illustrate the performance of the present hybrid method. Copyright © 2004 John Wiley & Sons, Ltd.