共查询到20条相似文献，搜索用时 125 毫秒
1.
《Information processing & management》2022,59(3):102891
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. 相似文献
2.
A specialized method is presented for listing all the spinning trees of the wheel, homeomorphs of the wheel, and certain cellular arrays. The procedure is a generalization of a known method of enumerating the trees of a suitably labeled ladder graph, and results in a direct listing of the trees with no duplications and no extraneous subgraphs. 相似文献
3.
Bond graphs are an extremely useful modeling procedure for representing the actual energy exchange mechanisms of interacting dynamic systems. Governing state equations are straightforwardly obtained from the bond graph; however, for large structures, a restrictively large number of equations can result. A procedure is developed whereby the original equations are reduced to a form suitable for modal decomposition. The resulting modes are reinterpreted in bond graph form with the resulting model being an extremely accurate system representation while requiring only a fraction of the original number of equations. The procedure is demonstrated through example. 相似文献
4.
In this paper, the distributed fault diagnosis (DFD) of networked dynamical systems with time-varying connected topologies, e.g., wireless sensor networks in harsh environments, is considered. Specifically, two essential problems are focused on, which are faced in extending the the decomposition-based adaptive DFD approach to such topology-varying systems. The problems introduced by the time-varying topologies are, respectively, decomposition schemes deterioration and pre-training difficulties. The causes of the two problems are detailed and addressed in our work. First, for the decomposition schemes deterioration problem, a multi-agent dynamics-based online distributed decomposition algorithm are developed, so that a decent decomposed network structure for such topology-varying network can be maintained. Second, to alleviate the pre-training difficulties in topology-varying systems, a fault detection method is proposed, which avoids the need for pre-training. The distributed decomposition algorithm is proved to converge in finite steps, and the proposed fault detection method is verified both theoretically and experimentally. 相似文献
5.
6.
《Journal of The Franklin Institute》2023,360(4):3307-3329
The problem of the reachable set (RS) control of sliding mode control (SMC) for a class of singular systems with or without time-varying delay under zero initial conditions is studied. The purpose is to get an RS boundary containing all states of the system by designing an SMC. Firstly, singular systems with or without time-varying delay are decomposed into slow and fast subsystems by using the decomposition approach. Then, the augmented Lyapunov functional is built utilizing the decomposed state vector. The SMC is designed based on the exponential reaching criterion, resulting in the corresponding closed-loop control system (CLCS) construction. As a consequence, an RS criterion is constructed by employing the inequality scaling approach and the free-weighting matrix in conjunction with the linear matrix inequality (LMI). Finally, the validity and primacy of the results are provided by two numerical and practical examples. 相似文献
7.
针对基于四象限压力传感器阵列的迎角测量方法中传感器信号受到干扰的问题,讨论了在其预处理中应用小波包变换进行信号分离提取的方法。根据小波包分解理论,对信号进行小波包分解,信号中频率不同的部分落在不同的尺度上,剔除反映干扰的变换尺度,提取出有用信息。结果表明,该方法可很好地提取信号的主要特征信息,也有效地去除了确定性干扰和随机噪声,与传统的信号滤波方法相比较具有明显的优点,同时为后续的信号处理奠定了基础。 相似文献
8.
The graph theoretic aspects of controllability and observability are examined and related to the tensorial formulation of Part I of the paper. Particular emphasis is given to the significance of the system digraph decomposition and the relevance of this to certain system algebraic properties of interest in control theory. 相似文献
9.
《Journal of The Franklin Institute》2023,360(4):2769-2793
This study focuses on the observability of second-order linear time invariant (LTI) systems with incommensurable output matrices through a matrix-weighted graph. Here, the observability of such systems refers to that the relative outputs have synchronized solutions for the identical LTI systems. Compared with most of existing results, relying on scalar networks (i.e., the weight of edges is a constant), this study investigates the observability in a matrix-weight-based network. Some necessary and sufficient conditions for the observability have been obtained by the space analysis, spectral analysis and matrix decomposition, respectively. Moreover, the relationship between the observability and the connectivity of its interconnection graph is also discussed. Examples and simulations are shown to verify the theoretical results. 相似文献
10.
Camilla B. Schwind 《Information processing & management》1983,19(4):223-235
This paper presents a formalism for the representation of complex semantic relations among concepts of natural language. We define a semantic algebra as a set of atomic concepts together with an ordered set of semantic relations. Semantic trees are a graphical representation of a semantic algebra (comparable to Kantorovic trees for boolean or arithmetical expressions). A semantic tree is an ordered tree with nodes labeled with relation and concept names. We generate semantic trees from natural language texts in such a way that they represent the semantic relations which hold among the concepts occurring within that text. This generation process is carried out by a transformational grammar which transforms directly natural language sentences into semantic trees. We present an example for concepts and relations within the domain of computer science where we have generated semantic trees from definition texts by means of a metalanguage for transformational grammars (a sort of metacompiler for transformational grammars). The semantic trees generated so far serve for thesaurus entries in an information retrieval system. 相似文献
11.
The problem is to determine the linear graph that has the maximum number of spanning trees, where only the number of nodes N and the number of branches B are prescribed. We deal with connected graphs G(N,B) obtained by deleting D branches from a complete graph KN. Our solution is for D less than or equal to N 相似文献
12.
This paper deals with the problem of finding graphs (directed and undirected) maximizing the number of spanning trees among the regular graphs with the same number of nodes and edges. The approach is based on heuristic algorithms such as k-optimal and evolutionary. The emphasis is rather on checking whether these techniques are applicable to solving extremal graph problems than investigating generic structures of optimal graphs. For this reason circulant graphs, for which computationally effective tree counting formulas exist, are discussed first and then the results extended to cover the class of regular graphs. 相似文献
13.
Yanping Gao Jingwei Ma Min Zuo Tongqiang Jiang Junping Du 《Journal of The Franklin Institute》2012,349(8):2598-2608
This paper studies the consensus problem of multiple agents with discrete-time second-order dynamics. It is assumed that the information obtained by each agent is with time-varying delays and the interaction topology is time-varying, where the associated direct graphs may not have spanning trees. Under the condition that the union graph is strongly connected and balanced, it is shown that there exist controller gains such that consensus can be reached for any bounded time-delays. Moreover, a method is provided to design controller gains. Simulations are performed to validate the theoretical results. 相似文献
14.
《Information processing & management》2022,59(4):102953
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. 相似文献
15.
Each graph may be associated with a certain function called its generic form. If one knows the generic forms of given graphs, then one can easily determine the number of spanning trees in graphs obtained from a complete multi-graph either (1) by adding, or (2) by deleting the edges of disjoint copies of the given graphs. Our obejective here is to give a proof of a simple and useful relation between the generic forms of graphs that are complementary with respect to a complete multi-graph. 相似文献
16.
In this paper, the issue about the stationary distribution for hybrid multi-stochastic-weight coupled networks (HMSWCN) via aperiodically intermittent control is investigated. Specially, when stochastic disturbance gets to zero, the exponential stability in pth moment for hybrid multi-weight coupled networks (HMWCN) is considered. Under the framework of the Lyapunov method, M-matrix and Kirchhoff’s Matrix Tree Theorem in the graph theory, several sufficient conditions are derived to guarantee the existence of a stationary distribution and exponential stability. Different from previous work, the existing area of a stationary distribution is not only related to the topological structure of coupled networks, but also aperiodically intermittent control (the rate of control width and control duration). Subsequently, as an application to theoretical results, a class of hybrid multi-stochastic-weight coupled oscillators is studied. Ultimately, numerical examples are carried out to demonstrate the effectiveness of theoretical results and effects of the control schemes. 相似文献
17.
《Information processing & management》2023,60(4):103387
Precise prediction of Multivariate Time Series (MTS) has been playing a pivotal role in numerous kinds of applications. Existing works have made significant efforts to capture temporal tendency and periodical patterns, but they always ignore abrupt variations and heterogeneous/spatial associations of sensory data. In this paper, we develop a dual normalization (dual-norm) based dynamic graph diffusion network (DNGDN) to capture hidden intricate correlations of MTS data for temporal prediction. Specifically, we design time series decomposition and dual-norm mechanism to learn the latent dependencies and alleviate the adverse effect of abnormal MTS data. Furthermore, a dynamic graph diffusion network is adopted for adaptively exploring the spatial correlations among variables. Extensive experiments are performed on 3 real world experimental datasets with 8 representative baselines for temporal prediction. The performances of DNGDN outperforms all baselines with at least 4% lower MAPE over all datasets. 相似文献
18.
Xian-He Zhang Guang-Song Han Zhi-Hong Guan Juan Li Ding-Xue Zhang Rui-Quan Liao 《Journal of The Franklin Institute》2018,355(8):3677-3690
A robust multi-tracking problem is solved for heterogeneous multi-agent systems with uncertain nonlinearities and disturbances. The nonlinear function satisfies a Lipschitz condition with a time-varying gain, the integral of which is bounded by a linear function. A distributed impulsive protocol is proposed, where the position data and velocity data of desired trajectories are needed only at sampling instants. Based on the system decomposition technique, the error dynamic system of achieving multi-tracking is decomposed into two impulsive dynamic systems with vanishing perturbation and nonvanishing perturbation, respectively. Constructing a nominal model, then the multi-tracking problem is converted into the stability of impulsive dynamic system with nonvanishing perturbation under some conditions. It is proved that the proposed impulsive protocol is robust enough to solve the multi-tracking problem. Numerical examples are presented to illustrate the effectiveness of our theoretical results. 相似文献
19.
《Information processing & management》2020,57(2):102052
Finding structural and efficient ways of leveraging available data is not an easy task, especially when dealing with network data, as is the case in telco churn prediction. Several previous works have made advancements in this direction both from the perspective of churn prediction, by proposing augmented call graph architectures, and from the perspective of graph featurization, by proposing different graph representation learning methods, frequently exploiting random walks. However, both graph augmentation as well as representation learning-based featurization face drawbacks. In this work, we first shift the focus from a homogeneous to a heterogeneous perspective, by defining different probabilistic meta paths on augmented call graphs. Secondly, we focus on solutions for the usually significant number of random walks that graph representation learning methods require. To this end, we propose a sampling method for random walks based on a combination of most suitable random walk generation strategies, which we determine with the help of corresponding Markov models. In our experimental evaluation, we demonstrate the benefits of probabilistic meta path-based walk generation in terms of predictive power. In addition, this paper provides promising insights regarding the interplay of the type of meta path and the predictive outcome, as well as the potential of sampling random walks based on the meta path structure in order to alleviate the computational requirements of representation learning by reducing typically sizable required data input. 相似文献
20.
中国省际全要素生产率的空间差异与变动趋势 总被引：1，自引：0，他引：1
本文采用DEA三阶段模型和Malmquist指数分解方法,对中国1987-2008年28个省市的全要素生产率进行测度和分解,分析了全要素生产率的空间差异和变动趋势,并利用Kernel密度分布法进行了要素投入和全要素生产率的经济差距效应分析。结果表明,环境因素确实对效率测度存在较大影响,DEA三阶段模型增强了效率评价的科学性和省市间的可比性;技术效率是影响全要素生产率变动的主要因素,全要素生产率的变动存在较大的空间差异,但呈缩小趋势;要素的空间均匀分布对缩小中国区域经济差距起到较大的贡献。 相似文献