Robust Computation of Intersection Graph between Two Solids

We propose a new robust algorithm for Boolean operations on solid models. The algorithm produces a consistent intersection graph between two input solids whose geometrical data are represented in floating point numbers. In order to prevent numerical calculation errors and inaccuracy of input data from causing inconsistency of the output, we put higher priority on symbolical connectivity of the edge-face intersection points than their numerical nearness. Each edge-face intersection point is symbolically represented using face names, which generate connectivity relations between the intersection points and the intersection line segments. The symbols with the same connectivity are made into clusters. The intersection line segments connected together at their end clusters form the intersection graph of two solids. Inconsistency of the connectivity of the clusters is detected and the intersection graph is corrected automatically. We describe the algorithm in detail for polyhedral solids, discuss extension to curves solids, and show its effectiveness by some examples of Boolean operations for two solids whose faces intersect at a very small angle.     

基于事件项语义图聚类的多文档摘要方法

基于事件的抽取式摘要方法一般首先抽取那些描述重要事件的句子,然后把它们重组并生成摘要。该文将事件定义为事件项以及与其关联的命名实体,并聚焦从外部语义资源获取的事件项语义关系。首先基于事件项语义关系创建事件项语义关系图并使用改进的DBSCAN算法对事件项进行聚类,接着为每类选择一个代表事件项或者选择一类事件项来表示文档集的主题,最后从文档抽取那些包含代表项并且最重要的句子生成摘要。该文的实验结果证明在多文档自动摘要中考虑事件项语义关系是必要的和可行的。     

Perceptual grouping of line features in 3-D space: a model-based framework

In this paper, we propose a novel model-based perceptual grouping algorithm for the line features of 3-D polyhedral objects. Given a 3-D polyhedral model, perceptual grouping is performed to extract a set of 3-D line segments which are geometrically consistent with the 3-D model. Unlike the conventional approaches, grouping is done in 3-D space in a model-based framework. In our unique approach, a decision tree classifier is employed for encoding and retrieving the geometric information of the 3-D model. A Gestalt graph is constructed by classifying input instances into proper Gestalt relations using the decision tree. The Gestalt graph is then decomposed into a few subgraphs, yielding appropriate groups of features. As an application, we suggest a 3-D object recognition system which can be accomplished by selecting a best-matched group. In order to evaluate the performance of the proposed algorithm, experiments are carried out on both synthetic and real scenes.     

融合语义信息与注意力的图神经网络推荐算法

基于图神经网络的推荐算法通过从图中获取知识生成节点的特征表示,提高了推荐结果的可解释性.然而,随着推荐系统原始数据规模的不断扩大,大量包含语义信息的文本数据没有得到有效利用.同时图神经网络在融合图中邻居信息时没有区分关键节点,使得模型难以学习到高质量的实体特征,进而导致推荐质量下降.本文将图神经网络与语义模型相结合,提出一种融合语义信息与注意力的图神经网络推荐算法.该算法基于SpanBERT语义模型处理实体相关的文本信息,生成包含语义信息的特征嵌入,并将注意力机制引入到基于用户社交关系以及用户-项目交互的影响传播融合过程中,从而实现用户和项目两类实体特征的有效更新.在公开数据集上的对比实验结果表明,本文所提出的方法较现有基准方法在各项指标上均有所提升.     

Abstractive Text Summarization based on Improved Semantic Graph Approach

The goal of abstractive summarization of multi-documents is to automatically produce a condensed version of the document text and maintain the significant information. Most of the graph-based extractive methods represent sentence as bag of words and utilize content similarity measure, which might fail to detect semantically equivalent redundant sentences. On other hand, graph based abstractive method depends on domain expert to build a semantic graph from manually created ontology, which requires time and effort. This work presents a semantic graph approach with improved ranking algorithm for abstractive summarization of multi-documents. The semantic graph is built from the source documents in a manner that the graph nodes denote the predicate argument structures (PASs)—the semantic structure of sentence, which is automatically identified by using semantic role labeling; while graph edges represent similarity weight, which is computed from PASs semantic similarity. In order to reflect the impact of both document and document set on PASs, the edge of semantic graph is further augmented with PAS-to-document and PAS-to-document set relationships. The important graph nodes (PASs) are ranked using the improved graph ranking algorithm. The redundant PASs are reduced by using maximal marginal relevance for re-ranking the PASs and finally summary sentences are generated from the top ranked PASs using language generation. Experiment of this research is accomplished using DUC-2002, a standard dataset for document summarization. Experimental findings signify that the proposed approach shows superior performance than other summarization approaches.     

标签集约束近似频繁模式的并行挖掘

近似频繁模式衍生于频繁模式,综合了频繁项集与频繁子图的特点。针对该模式的研究集中在无标签图上,其应用场景主要为社交网络、语义网络、智能电网等。近似频繁模式挖掘过程同时涉及频繁项集挖掘和频繁子图挖掘,因此已有的处理频繁模式挖掘算法无法较好地解决近似频繁模式挖掘问题。基于近似频繁模式结构,将其拓展到带标签图中,引入标签集约束,并设计标签集约束近似频繁模式挖掘算法LCPP（Label-Constraint Proximity Pattern）,该算法并行部署在MapReduce计算模型中,弥补了开源pFP算法处理大规模数据时效率不高的缺点。实验结果验证了该算法的有效性和可扩展性,表明了LCPP算法是pFP算法的极佳补充。     

基于时序结构图的视频流描述方法

通过对视频流的分解可以获得基于关键帧集的视频流表示,但这种表示方法不能反映出视频流中隐藏的故事发展关系,为揭示这种关系,提出了一种视频流的快速聚类算法,用于对视频流分解单元进行相关性分析,该算法通过检测视频镜头间的相似性和连续性,实现把来自同一摄像机的视频镜头归并入同一视频类,并帱此得到而且为矿山频流的快速浏览和检索提供了新的思路。     

Large scale multiple robot visual mapping with heterogeneous landmarks in semi-structured terrain

This paper addresses the cooperative localization and visual mapping problem with multiple heterogeneous robots. The approach is designed to deal with the challenging large semi-structured outdoors environments in which aerial/ground ensembles are to evolve. We propose the use of heterogeneous visual landmarks, points and line segments, to achieve effective cooperation in such environments. A large-scale SLAM algorithm is generalized to handle multiple robots, in which a global graph maintains the relative relationships between a series of local sub-maps built by the different robots. The key issue when dealing with multiple robots is to find the link between them, and to integrate these relations to maintain the overall geometric consistency; the events that introduce these links on the global graph are described in detail. Monocular cameras are considered as the primary extereoceptive sensor. In order to achieve the undelayed initialization required by the bearing-only observations, the well-known inverse-depth parametrization is adopted to estimate 3D points. Similarly, to estimate 3D line segments, we present a novel parametrization based on anchored Plücker coordinates, to which extensible endpoints are added. Extensive simulations show the proposed developments, and the overall approach is illustrated using real-data taken with a helicopter and a ground rover.     

A metric for line segments

This correspondence presents a metric for describing line segments. This metric measures how well two line segments can be replaced by a single longer one. This depends for example on collinearity and nearness of the line segments. The metric is constructed using a new technique using so-called neighborhood functions. The behavior of the metric depends on the neighborhood function chosen. In this correspondence, an appropriate choice for the case of line segments is presented. The quality of the metric is verified by using it in a simple clustering algorithm that groups line segments found by an edge detection algorithm in an image. The fact that the clustering algorithm can detect long linear structures in an image shows that the metric is a good measure for the groupability of line segments     

Building spatial temporal relation graph of concepts pair using web repository

Mining semantic relations between concepts underlies many fundamental tasks including natural language processing, web mining, information retrieval, and web search. In order to describe the semantic relation between concepts, in this paper, the problem of automatically generating spatial temporal relation graph (STRG) of semantic relation between concepts is studied. The spatial temporal relation graph of semantic relation between concepts includes relation words, relation sentences, relation factor, relation graph, faceted feature, temporal feature, and spatial feature. The proposed method can automatically generate the spatial temporal relation graph (STRG) of semantic relation between concepts, which is different from the manually generated annotation repository such as WordNet and Wikipedia. Moreover, the proposed method does not need any prior knowledge such as ontology or the hierarchical knowledge base such as WordNet. Empirical experiments on real dataset show that the proposed algorithm is effective and accurate.     

基于语义属性特征图的查询扩展方法

传统的查询扩展方法由于忽略了词之间的语义关系,在不规范的短小关键字上补充扩展的词已经无法达到预期目标。Linked Data技术利用资源描述框架(RDF)图模型形成Linked Open Data Cloud,能提供更多语义信息。针对查询扩展忽略语义的问题,提出了一种基于语义属性特征图的查询扩展方法。该方法将语义网与图的思想融合,利用以DBpedia资源为顶点的属性图加以扩展。首先,通过有监督的学习训练出15种语义属性特征的权重,用于表达扩展资源的有用性;然后,在整个DBpedia图上通过标签属性实现查询关键字到DBpedia匹配资源的映射;再根据属性特征广度搜索出邻接点,并将其作为扩展候选词,最后筛选出词相关行分值最高的作为最终扩展词。实验表明,与LOD Keyword Expansion方法相比,基于语义属性特征图的扩展方法召回率达到0.89,平均逆排序(MRR)提高4个百分点,与用户查询更匹配。     

基于知网和知识图的汉语词语语义相似度算法

汉语词语间语义相似是词语间的基本关系之一,文章提出了一种基于知网和知识图的词语语义相似度计算的方法,通过改进传统的知识图表示方式,根据知网中概念项的抽取结果对词语的义项进行表示,用词图的相似度来表示相应词语的语义相似度。实验结果表明该算法对词语间语义相似度计算是有效的。     

Retrieval by classification of images containing large manmade objects using perceptual grouping

This paper applies perceptual grouping rules to the retrieval by classification of images containing large manmade objects such as buildings, towers, bridges, and other architectural objects. The semantic interrelationships between primitive image features are exploited by perceptual grouping to extract structure to detect the presence of manmade objects. Segmentation and detailed object representation are not required. The system analyzes each image to extract features that are strong evidence of the presence of these objects. These features are generated by the strong boundaries typical of manmade structures: straight line segments, longer linear lines, coterminations, “L” junctions, “U” junctions, parallel lines, parallel groups, “significant” parallel groups, cotermination graph, and polygons. A K-nearest neighbor framework is employed to classify these features and retrieve the images that contain manmade objects. Results are demonstrated for two databases of monocular outdoor images.     

Context‐based email classification model

Context‐based email classification requires understanding of semantic and structural attributes of email. Most of the research has focused on generating semantic properties through structural components of email. By viewing emails as events (as a major subset of class of email), a rich contextual test‐bed representation for understanding of the semantic attributes of emails has been devised. The event‐ based emails have traditionally been studied based on simple structural properties. In this paper, we present a novel approach by first representing such class of emails as graphs, followed by heuristically applying graph mining and matching algorithm to pick templates representing contextual and semantic attributes that help classify emails. The classification templates used three key event classes: social, personal and professional. Results show that our graph mining and matching supported template‐based approach performs consistently well over event email data set with high accuracy.     

A graph grammar approach to geographical databases

This paper treats the modeling of an important class of databases, i.e. geographical databases, with emphasis on both structural (data definition) and behavioral (data manipulation) aspects. Geometric objects such as polygons, line segments, and points may have different relations among each other (such as order, adjacency, connectivity) and can be represented in a uniform spatial data structure (structure graph). The dynamic behavior is defined by a finite set of consistency-preserving state transitions (productions) where coincidence problems as well as topological properties have to be solved. Moreover, the graph grammar approach can be used to study the synchronization of several concurrent productions (Church-Rosser properties) and offers a framework for implementing a geographical database.     

Globally optimal grouping for symmetric closed boundaries by combining boundary and region information

Many natural and man-made structures have a boundary that shows a certain level of bilateral symmetry, a property that plays an important role in both human and computer vision. In this paper, we present a new grouping method for detecting closed boundaries with symmetry. We first construct a new type of grouping token in the form of symmetric trapezoids by pairing line segments detected from the image. A closed boundary can then be achieved by connecting some trapezoids with a sequence of gap-filling quadrilaterals. For such a closed boundary, we define a unified grouping cost function in a ratio form: the numerator reflects the boundary information of proximity and symmetry and the denominator reflects the region information of the enclosed area. The introduction of the region-area information in the denominator is able to avoid a bias toward shorter boundaries. We then develop a new graph model to represent the grouping tokens. In this new graph model, the grouping cost function can be encoded by carefully designed edge weights and the desired optimal boundary corresponds to a special cycle with a minimum ratio-form cost. We finally show that such a cycle can be found in polynomial time using a previous graph algorithm. We implement this symmetry-grouping method and test it on a set of synthetic data and real images. The performance is compared to two previous grouping methods that do not consider symmetry in their grouping cost functions.     

一种基于语义关系图的词语语义相关度计算模型

词语的语义计算是自然语言处理领域的重要问题之一,目前的研究主要集中在词语语义的相似度计算方面,对词语语义的相关度计算方法研究不够.为此,本文提出了一种基于语义词典和语料库相结合的词语语义相关度计算模型.首先,以HowNet和大规模语料库为基础,制定了相关的语义关系提取规则,抽取了大量的语义依存关系;然后,以语义关系三元组为存储形式,构建了语义关系图;最后,采用图论的相关理论,对语义关系图中的语义关系进行处理,设计了一个基于语义关系图的词语语义相关度计算模型.实验结果表明,本文提出的模型在词语语义相关度计算方面具有较好的效果,在WordSimilarity-353数据集上的斯皮尔曼等级相关系数达到了0.5358,显著地提升了中文词语语义相关度的计算效果.     

基于知识图的汉语基本名词短语分析模型

本文提出了一种基于知识图的汉语baseNP分析模型。它以知识图为知识表示方法,利用《知网》为语义知识资源,采用以语义为主、语法为辅的策略,先为短语中的每一个实词构造“词图”,然后合并“词图”而组成“短语图”,最后得到一个关于汉语baseNP结构信息和语义信息的知识图。因此它不仅分析了汉语baseNP结构的内部句法关系,而且分析了汉语baseNP结构成分间的语义关系并以知识图的形式表示出了这种语义关系。实验结果表明这个模型对于汉语baseNP的分析是有效的。     

全局显著结构主导下的知觉编组算法

提出一种建立在可靠的全局线索基础上的编组算法.编组线索为反映全局显著结构的拓扑特征闭合性和平行性以及局部规律邻接性和连续性.依据概率推理选择最显著的边缘作为种子,依据全局依赖性选择最有可能与种子属于同一编组的边缘.编组的形成中融入注意机制,一方面缩小寻优空间另一方面确定各编组被检测的顺序.在Berkley图像库上的实验表明,该算法至少具有与Ncut和mini-cut相当的准确率,特别对纹理少的图像能够有效地降低错编率与漏编率.同时由于对边缘进行编组降低了输人数据的维数,因此比Ncut和mini-cut更少地受到图像尺寸的限制.