Structure-sensitive semantic matching for aggregate question answering over knowledge base

Aggregate question answering essentially returns answers for given questions by obtaining query graphs with unique dependencies between values and corresponding objects. Word order dependency, as the key to uniquely identify dependency of the query graph, reflects the dependencies between the words in the question. However, due to the semantic gap caused by the expression difference between questions encoded with word vectors and query graphs represented with logical formal elements, it is not trivial to match the correct query graph for the question. Most existing approaches design more expressive query graphs for complex questions and rank them just by directly calculating their similarities, ignoring the semantic gap between them. In this paper, we propose a novel Structure-sensitive Semantic Matching(SSM) approach that learns aligned representations of dependencies in questions and query graphs to eliminate their gap. First, we propose a cross-structure matching module to bridge the gap between two modalities(i.e., textual question and query graph). Then, we propose an entropy-based gated AQG filter to remove the structural noise caused by the uncertainty of dependencies. Finally, we present a two-channel query graph representation that fuses the semantics of abstract structure and grounding content of the query graph explicitly. Experimental results show that SSM could learn aligned representations of questions and query graphs to eliminate the gaps between their dependencies, and improves up to 12% (F1 score) on aggregation questions of two benchmark datasets.     

A Case Report of Germline Compound Heterozygous Mutations in the BRCA1 Gene of an Ovarian and Breast Cancer Patient

The germline carrier of the BRCA1 pathogenic mutation has been well proven to confer an increased risk of breast and ovarian cancer. Despite BRCA1 biallelic pathogenic mutations being extremely rare, they have been reported to be embryonically lethal or to cause Fanconi anemia (FA). Here we describe a patient who was a 48-year-old female identified with biallelic pathogenic mutations of the BRCA1 gene, with no or very subtle FA-features. She was diagnosed with ovarian cancer and breast cancer at the ages of 43 and 44 and had a strong family history of breast and gynecological cancers.     

Generating Questions Based on Semi-Automated and End-to-End Neural Network

With the emergence of large-scale knowledge base, how to use triple information to generate natural questions is a key technology in question answering systems. The traditional way of generating questions require a lot of manual intervention and produce lots of noise. To solve these problems, we propose a joint model based on semi-automated model and End-to-End neural network to automatically generate questions. The semi-automated model can generate question templates and real questions combining the knowledge base and center graph. The End-to-End neural network directly sends the knowledge base and real questions to BiLSTM network. Meanwhile, the attention mechanism is utilized in the decoding layer, which makes the triples and generated questions more relevant. Finally, the experimental results on SimpleQuestions demonstrate the effectiveness of the proposed approach.     

中医学视角下城市物质空间的生命要素探析

通过借鉴中医学整体观和生命观的相关理论,从认识城市有机体的生命属性入手,发现并提炼能够产生和传递城市运营所需各种能量的生命要素:廊道和功能体。阐释了其多层级、多性质和多形态的特点,提出了功能体有动力、廊道成系统、敏感点有活力的疏通策略,旨在促进城市各种能量的有机循环,从而维护城市的健康与安全,提高城市生活的效率与质量。     

基于桥连接的词典学习方法的语义解析

现阶段的语义解析方法大部分都基于组合语义,这类方法的核心就是词典。词典是词汇的集合,词汇定义了自然语言句子中词语到知识库本体中谓词的映射。语义解析一直面临着词典中词汇覆盖度不够的问题。针对此问题,该文在现有工作的基础上,提出了基于桥连接的词典学习方法,该方法能够在训练中自动引入新的词汇并加以学习,为了进一步提高新学习到的词汇的准确度,该文设计了新的词语—二元谓词的特征模板,并使用基于投票机制的核心词典获取方法。该文在两个公开数据集(WebQuestions和Free917)上进行了对比实验,实验结果表明,该文方法能够学习到新的词汇,提高词汇的覆盖度,进而提升语义解析系统的性能,特别是召回率。     

Indoor solid fuel use for heating and cooking with blood pressure and hypertension: A cross-sectional study among middle-aged and older adults in China

A cross-sectional study was conducted to investigate the impact of solid fuel use for heating and cooking on blood pressure (BP) and hypertension, using data from the China Health and Retirement Longitudinal Study (CHARLS). The primary fuels used for indoor heating and cooking were collected by questionnaires, respectively. Hypertension was defined based on self-report of physician's diagnosis, and/or measured BP, and/or anti-hypertensive medication use. Multivariate logistic regression models were constructed to assess the associations. Among 10 450 eligible participants, 68.2% and 57.2% used indoor solid fuel for heating and cooking, respectively. Compared with none/clean fuel users, solid fuel for heating was associated with elevated BP (adjusted β: 2.02, 95% CI: 1.04–3.01 for systolic BP; adjusted β: 1.36, 95% CI: 0.78–1.94 for diastolic BP) and increased risk of hypertension (adjusted odds ratio: 1.15, 95% CI: 1.03–1.29). The impact of indoor solid fuel for heating on BP was more evident in rural and north residents, and hypertensive patients. We did not detect any significant associations between solid fuel use for cooking and BP/hypertension. Indoor solid fuel use is prevalent in China, especially in the rural areas. Its negative impact on BP suggested that modernization of household fuel use may help to reduce the burden of hypertension in China.     

M-estimator based Chinese Remainder Theorem with few remainders using a Kroenecker product based mapping vector

The Chinese Remainder Theorem (CRT) explains how to estimate an integer-valued number from the knowledge of the remainders obtained by dividing such unknown integer by co-prime integers. As an algebraic theorem, CRT is the basis for several techniques concerning data processing. For instance, considering a single-tone signal whose frequency value is above the sampling rate, the respective peak in the DFT informs the impinging frequency value modulo the sampling rate. CRT is nevertheless sensitive to errors in the remainders, and many efforts have been developed in order to improve its robustness. In this paper, we propose a technique to estimate real-valued numbers by means of CRT, employing for this goal a Kroenecker based M-Estimation (ME), specially suitable for CRT systems with low number of remainders. Since ME schemes are in general computationally expensive, we propose a mapping vector obtained via Kroenecker products which considerably reduces the computational complexity. Furthermore, our proposed technique enhances the probability of estimating an unknown number accurately even when the errors in the remainders surpass 1/4 of the greatest common divisor of all moduli. We also provide a version of the mapping vectors based on tensorial n-mode products, delivering in the end the same information of the original method. Our approach outperforms the state-of-the-art CRT methods not only in terms of percentage of successful estimations but also in terms of smaller average error.     

Precipitating spinel into precursor glass and its assistance in crystallization

The crystallization phenomena of spinel in CaO-MgO-Al₂O₃-SiO₂-Fe₂O₃ glass have received much attention due to the particular role in preparation of glass-ceramic materials, which represent an effective option to manage hazardous waste. In this study, both preliminary spinel and secondary spinel were precipitated in the precursor glass. The formation of these spinel was meticulously assessed by a combination of X-ray diffractometry and scanning electron microscopy. The structure of the microenvironment in the precursor glass was characterized by Raman spectrums. These advanced techniques highlight the potential for one-step crystallization of the glass. The investigation, which focused on one-step crystallization, demonstrated the growth of pyroxene on spinel accompanying a migration of chromium. The results also show the microstructure of the obtained glass-ceramic was very dependent on the heat-treat temperature. This study not only unambiguously reveals the precipitation mechanisms of spinel but also provides more documentation for one-step crystallization in the glass-ceramics field.     

基于特征融合的中文简历解析方法研究

针对基于规则和统计的传统中文简历解析方法效率低、成本高、泛化能力差的缺点，提出一种基于特征融合的中文简历解析方法，即级联Word2Vec生成的词向量和用BLSTM（Bidirectional Long Short-Term Memory）建模字序列生成的词向量，然后再结合BLSTM和CRF（Conditional Random Fields）对中文简历进行解析（BLSTM-CRF）。为了提高中文简历解析的效率，级联包含字序列信息的词向量和用Word2Vec生成的词向量，融合成一个新的词向量表示；再由BLSTM强大的学习能力融合词的上下文信息，输出所有可能标签序列的分值给CRF层；再由CRF引入标签之间约束关系求解最优序列。利用梯度下降算法训练神经网络，使用预先训练的词向量和Dropout优化神经网络，最终完成对中文简历的解析工作。实验结果表明，所提的特征融合方法优于传统的简历解析方法。     

移动信息接收机的研制

介绍了一种新型的通信传媒设备，详细介绍了该接收机的工作机理、结构、功能特点以及软硬件的设计思路。