Incorporating rich syntax information in Grammatical Error Correction

Syntax parse trees are a method of representing sentence structure and are often used to provide models with syntax information and enhance downstream task performance. Because grammar and syntax are inherently linked, the incorporation of syntax parse trees in GEC is a natural solution. In this work, we present a method of incorporating syntax parse trees for Grammatical Error Correction (GEC). Building off a strong sequence-to-sequence Transformer baseline, we present a unified parse integration method for GEC that allows for the use of both dependency and constituency parse trees, as well as their combination - a syntactic graph. Specifically, on the sentence encoder, we propose a graph encoder that can encode dependency trees and constituent trees at the same time, yielding two representations for terminal nodes (i.e., the token of the sentence) and non-terminal nodes. We next use two cross-attentions (NT-Cross-Attention and T-Cross-Attention) to aggregate these source syntactic representations to the target side for final corrections prediction. In addition to evaluating our models on the popular CoNLL-2014 Shared Task and JFLEG GEC benchmarks, we affirm the effectiveness of our proposed method by testing both varying levels of parsing quality and exploring the use of both parsing formalisms. With further empirical exploration and analysis to identify the source of improvement, we found that rich syntax information provided clear clues for GEC; a syntactic graph composed of multiple syntactic parse trees can effectively compensate for the limited quality and insufficient error correction capability of a single syntactic parse tree.     

Interest Evolution-driven Gated Neighborhood aggregation representation for dynamic recommendation in e-commerce

Recommender system as an effective method to reduce information overload has been widely used in the e-commerce field. Existing studies mainly capture semantic features by considering user-item interactions or behavioral history records, which ignores the sparsity of interactions and the drift of user preferences. To cope with these challenges, we introduce the recently popular Graph Neural Networks (GNN) and propose an Interest Evolution-driven Gated Neighborhood (IEGN) aggregation representation model which can capture accurate user representation and track the evolution of user interests. Specifically, in IEGN, we explicitly model the relational information between neighbor nodes by introducing the gated adaptive propagation mechanism. Then, a personalized time interval function is designed to track the evolution of user interests. In addition, a high-order convolutional pooling operation is used to capture the correlation among the short-term interaction sequence. The user preferences are predicted by the fusion of user dynamic preferences and short-term interaction features. Extensive experiments on Amazon and Alibaba datasets show that IEGN outperforms several state-of-the-art methods in recommendation tasks.     

Aspect sentiment analysis with heterogeneous graph neural networks

Aspect-based sentiment analysis technologies may be a very practical methodology for securities trading, commodity sales, movie rating websites, etc. Most recent studies adopt the recurrent neural network or attention-based neural network methods to infer aspect sentiment using opinion context terms and sentence dependency trees. However, due to a sentence often having multiple aspects sentiment representation, these models are hard to achieve satisfactory classification results. In this paper, we discuss these problems by encoding sentence syntax tree, words relations and opinion dictionary information in a unified framework. We called this method heterogeneous graph neural networks (Hete_GNNs). Firstly, we adopt the interactive aspect words and contexts to encode the sentence sequence information for parameter sharing. Then, we utilized a novel heterogeneous graph neural network for encoding these sentences’ syntax dependency tree, prior sentiment dictionary, and some part-of-speech tagging information for sentiment prediction. We perform the Hete_GNNs sentiment judgment and report the experiments on five domain datasets, and the results confirm that the heterogeneous context information can be better captured with heterogeneous graph neural networks. The improvement of the proposed method is demonstrated by aspect sentiment classification task comparison.     

中国创新合作网络的演进特征及影响因素研究——以SCI论文合作为例

创新驱动战略实施背景下,融入创新合作网络是中国各省域提升区域创新能力的重要途径。本文基于2008—2018年中国省际之间在Web of Science核心合集发表的论文合作数,构建中国省际创新合作网络,描绘其时空演化特征,并借助零膨胀负二项回归深入分析影响机制。研究发现：中国省际创新合作网络结构逐渐复杂化、均衡化,省际创新合作网络中网络节点之间的联系不断丰富,在网络中重要节点省份逐渐增多,而且随着时间推移,网络结构不断优化。多维邻近性检验显示网络邻近性、产业邻近性是影响创新合作关系的重要因素,地理邻近性、经济邻近性影响较小,网络邻近性可以通过调节地理邻近性、经济邻近性影响创新合作。     

专利联盟对遭受联盟伙伴专利诉讼的影响机制

    专利联盟能否帮助联盟成员防范联盟伙伴专利诉讼？基于联盟学习、合作竞争和社会网络理论,着眼于专利联盟网络,以2006—2018年MPEGLA管理的12个专利联盟成员为样本,通过14 454组配对,采用负二项回归模型实证研究专利联盟对遭受联盟伙伴专利诉讼的影响机制,并分析探索式合作、竞争性学习、网络中心位置和技术相似性的调节作用。研究发现：专利联盟与遭受联盟伙伴专利诉讼强度呈正向关系,探索式合作、竞争性学习和网络中心位置显著削弱专利联盟与专利诉讼的关系,而技术相似性显著强化专利联盟与专利诉讼的关系。本文拓展了专利联盟对联盟伙伴互动的影响路径,对于高技术企业有效利用专利联盟并防范联盟伙伴专利诉讼具有现实指导意义。     

利用K-shell分析合著网络中的作者传播影响力

以情报学领域的12种期刊在2000-2009年间的7 389位作者形成的合著网络为例,分别基于度和K-shell,介数和K-shell对作者传播影响力进行比较分析。结果表明,K-shell值较度、介数能更好地表征作者的传播影响力。这种方法可以推广到基于科技文献数据的其他网络中,如识别文献共被引网络、文献耦合网络中最具传播影响力的关键文献。     

基于图嵌入的特征提取算法研究

通过对图、完全图和正则图概念的介绍,详细地描述了图嵌入的方法,同时对主成分分析、线性鉴别分析、局部保持投影、保持近邻嵌入、L1图及其嵌入等经典的特征提取算法进行了详细的代数推导,列出了详细的推导过程,得出这些经典算法可以用图嵌入理论来解释的结论,最后得出特征提取算法的核心在于算法的图构造.     

Testing a communication network model of employee turnover based on centrality

This study tested Feeley and Barnett's (1997) Erosion Model (EM) of employee turnover which predicts that individuals who are more central in their communication network will be more likely to remain at their position (or less likely to turnover). Seventy employees from three different organizations were surveyed about their attitudes toward their jobs and were also asked to indicate (by checklist) which employees they spoke to regularly at work. Turnover data were obtained at 3 and 6 months time after the surveys were completed. Results generally supported the Erosion Model of employee turnover. Those employees with high Degree or number of links in the network were less likely to turnover. Employees who required fewer links to communicate to all others in the network (i.e., Closeness) were also less likely to turnover but this relationship only approached statistical significance (p = .06). Betweenness, defined as the frequency with which a person falls between pairs of other positions in a network, was also significantly related to employee turnover. It was also predicted, based on Feeley and Bamett's EM, that the relationship between network position and turnover would be mediated by an employee's level of commitment to the organization and his or her intentions to leave work. Closeness significantly predicted commitment while Betweenness and Degree were unrelated to commitment levels. Organizational commitment was negatively related to intentions to leave work and, unexpectedly, commitment levels were positively related to employee turnover. The results were discussed and the applications of this research for management practitioners were considered.     

方程、真值表与状态转换图在逻辑推理中的应用

给出了方程、真值表与状态转换图在逻辑推理中的应用。     

Petersen图类的边符号控制数

设图G=G(V,E),令函数f:E→{-1,1},f的权w(f)=∑x∈Ef[x],对x∈E中任一元素,定义f[x]=∑y∈N[x]f(y),这里N[x]表示E中x及其关联边的集合.图G的边符号控制函数为f:E→{-1,1},满足对所有的x∈E有f[x]≥1,图G的边符号控制数γS(G)就是图G上边符号控制数的最小权,称其f为图G的γS-函数.本文得到了Petersen图类的边符号控制数.