基于领域知识图谱的短文本实体链接

实体链接任务是识别文本中潜在的实体指称,并将其链接到给定知识库中无歧义的实体上。在绝大多数情况下,实体链接可能存在中文短文本缺乏有效上下文信息,导致存在一词多义的歧义现象;同时候选链接过程中,候选实体的不确定相关性也影响候选实体链接精确性。针对上述两个问题,提出深度神经网络与关联图相结合的实体链接模型。模型添加字符特征、上下文、信息深层语义来增强指称和实体表示,并进行相似度匹配。利用Fast-newman算法将图谱知识库聚类划分不同类型实体簇,将相似度计算得分最高候选实体所属实体簇映射到关系平面,构建聚类实体关联图。利用偏向随机游走算法考查候选实体之间语义相关度,计算指称与候选实体的匹配程度,输入链接实体。该模型可以实现短文本到知识图谱目标实体的准确链接。     

Disambiguating named entities with deep supervised learning via crowd labels

Named entity disambiguation (NED) is the task of linking mentions of ambiguous entities to their referenced entities in a knowledge base such as Wikipedia. We propose an approach to effectively disentangle the discriminative features in the manner of collaborative utilization of collective wisdom (via human-labeled crowd labels) and deep learning (via human-generated data) for the NED task. In particular, we devise a crowd model to elicit the underlying features (crowd features) from crowd labels that indicate a matching candidate for each mention, and then use the crowd features to fine-tune a dynamic convolutional neural network (DCNN). The learned DCNN is employed to obtain deep crowd features to enhance traditional hand-crafted features for the NED task. The proposed method substantially benefits from the utilization of crowd knowledge (via crowd labels) into a generic deep learning for the NED task. Experimental analysis demonstrates that the proposed approach is superior to the traditional hand-crafted features when enough crowd labels are gathered.     

多源异构数据的实体匹配方法研究

近年来,针对多源异构数据的实体匹配问题,已经有诸多学者提出不同的解决方法。然而,这些方法几乎都集中在RDFS或OWL等语义框架下进行实体匹配,不具有通用性。此外,针对多数据源实体匹配问题,目前主流解决方式是将其转换为多组两两数据源的实体匹配问题,该种方式直接进行两两匹配的计算复杂度过高,且没有从多数据源全局的角度分析问题。从这些问题出发,提出了一种的实体匹配方法,利用了实体中普遍存在的名称、属性和上下文信息,构建多种索引,缩减计算空间同时生成高质量的候选集;还定义了度量实体相似度的计算方法,有效地判别了实体对是否匹配。并根据实体间边的权重以及互斥关系,提出一种基于图划分的优化算法,划分多个等价实体构成的集合。从互联网中抓取商业领域下品牌和人物类别的真实数据进行实验测试,实验结果表明该方法取得了良好的效果。     

Logo localization and recognition in natural images using homographic class graphs

We propose a method for localization and classification of brand logos in natural images. The system has to overcome multiple challenges such as perspective deformations, warping, variations of the shape and colors, occlusions, background variations. To deal with perspective variation, we rely on homography matching between the SIFT keypoints of logo instances of the same class. To address the changes in color, we construct a weighted graph of logo interconnections that is further analyzed to extract potentially multiple instances of the class. The main instance is built by grouping the keypoints of the graph connected logos onto the central image. The secondary instance is needed for color inverted logos and is obtained by inverting the orientation of the main instance. The constructed logo recognition system is tested on two databases (FlickrLogos-32 and BelgaLogos), outperforming state of the art with more than 10 % accuracy.     

CoGCN: Combining co‐attention with graph convolutional network for entity linking with knowledge graphs

Entity linking is a fundamental task in natural language processing. The task of entity linking with knowledge graphs aims at linking mentions in text to their correct entities in a knowledge graph like DBpedia or YAGO2. Most of existing methods rely on hand‐designed features to model the contexts of mentions and entities, which are sparse and hard to calibrate. In this paper, we present a neural model that first combines co‐attention mechanism with graph convolutional network for entity linking with knowledge graphs, which extracts features of mentions and entities from their contexts automatically. Specifically, given the context of a mention and one of its candidate entities' context, we introduce the co‐attention mechanism to learn the relatedness between the mention context and the candidate entity context, and build the mention representation in consideration of such relatedness. Moreover, we propose a context‐aware graph convolutional network for entity representation, which takes both the graph structure of the candidate entity and its relatedness with the mention context into consideration. Experimental results show that our model consistently outperforms the baseline methods on five widely used datasets.     

融合多特征的中文集成实体链接方法

实体链接技术是将文本中的实体指称项正确链接到知识库中实体对象的过程，对知识库扩容起着关键作用。针对传统的实体链接方法主要利用上下文相似度等表层特征，而且忽略共现实体间的语义相关性，提出一种融合多特征的集成实体链接方法。首先结合同义词表、同名词表产生候选实体集，然后从多角度抽取语义特征，并将语义特征融合到构建的实体相关图中，最后对候选实体排序，选取top1实体作为链接目标。在NLP&CC2013中文微博实体链接评测数据集上进行实验，获得90.97%的准确率，与NLP&CC2013中文微博实体链接评测的最优系统相比，本文系统具有一定的优势。     

Answering pattern match queries in large graph databases via graph embedding

The growing popularity of graph databases has generated interesting data management problems, such as subgraph search, shortest path query, reachability verification, and pattern matching. Among these, a pattern match query is more flexible compared with a subgraph search and more informative compared with a shortest path or a reachability query. In this paper, we address distance-based pattern match queries over a large data graph G. Due to the huge search space, we adopt a filter-and-refine framework to answer a pattern match query over a large graph. We first find a set of candidate matches by a graph embedding technique and then evaluate these to find the exact matches. Extensive experiments confirm the superiority of our method.     

Aggregate nearest neighbor queries in uncertain graphs

Most recently, uncertain graph data begin attracting significant interests of database research community, because uncertainty is the intrinsic property of the real-world and data are more suitable to be modeled as graphs in numbers of applications, e.g. social network analysis, PPI networks in biology, and road network monitoring. Meanwhile, as one of the basic query operators, aggregate nearest neighbor (ANN) query retrieves a data entity whose aggregate distance, e.g. sum, max, to the given query data entities is smaller than those of other data entities in a database. ANN query on both certain graph data and high dimensional data has been well studied by previous work. However, existing ANN query processing approaches cannot handle the situation of uncertain graphs, because topological structures of an uncertain graph may vary in different possible worlds. Motivated by this, we propose the aggregate nearest neighbor query in uncertain graphs (UG-ANN) in this paper. First of all, we give the formal definition of UG-ANN query and the basic UG-ANN query algorithm. After that, to improve the efficiency of UG-ANN query processing, we develop two kinds of pruning approaches, i.e. structural pruning and instance pruning. The structural pruning takes advantages the monotonicity of the aggregate distance to derive the upper and lower bounds of the aggregate distance for reducing the graph size. Whereas, the instance pruning decreases the number of possible worlds to be checked in the searching tree. Comprehensive experimental results on real-world data sets demonstrate that the proposed method significantly improves the efficiency of the UG-ANN query processing.     

SPHeRe

The abundance of semantically related information has resulted in semantic heterogeneity. Ontology matching is among the utilized techniques implemented for semantic heterogeneity resolution; however, ontology matching being a computationally intensive problem can be a time-consuming process. Medium to large-scale ontologies can take from hours up to days of computation time depending upon the utilization of computational resources and complexity of matching algorithms. This delay in producing results, makes ontology matching unsuitable for semantic web-based interactive and semireal-time systems. This paper presents SPHeRe, a performance-based initiative that improves ontology matching performance by exploiting parallelism over multicore cloud platform. Parallelism has been overlooked by ontology matching systems. SPHeRe avails this opportunity and provides a solution by: (i) creating and caching serialized subsets of candidate ontologies with single-step parallel loading; (ii) lightweight matcher-based and redundancy-free subsets result in smaller memory footprints and faster load time; and (iii) implementing data parallelism based distribution over subsets of candidate ontologies by exploiting the multicore distributed hardware of cloud platform for parallel ontology matching and execution. Performance evaluation of SPHeRe on a trinode (12-core) private cloud infrastructure has shown up to 3 times faster ontology load time with up to 8 times smaller memory footprint than Web Ontology Language (OWL) frameworks Jena and OWLAPI. Furthermore, by utilizing the computation resources most efficiently, SPHeRe provides the best scalability in contrast with other ontology matching systems, i.e., GOMMA, LogMap, AROMA, and AgrMaker. On a private cloud instance with 8 cores, SPHeRe outperforms the most performance efficient ontology matching system GOMMA by 40 % in scalability and 4 times in performance.     

Efficient indexing for semantic search

The increasing performance and wider spread use of automated semantic annotation and entity linking platforms has empowered the possibility of using semantic information in information retrieval. While keyword-based information retrieval techniques have shown impressive performance, the addition of semantic information can increase retrieval performance by allowing for more accurate sense disambiguation, intent determination, and instance identification, just to name a few. Researchers have already delved into the possibility of integrating semantic information into practical search engines using a combination of techniques such as using graph databases, hybrid indices and adapted inverted indices, among others. One of the challenges with the efficient design of a search engine capable of considering semantic information is that it would need to be able to index information beyond the traditional information stored in inverted indices, including entity mentions and type relationships. The objective of our work in this paper is to investigate various ways in which different data structure types can be adopted to integrate three types of information including keywords, entities and types. We will systematically compare the performance of the different data structures for scenarios where (i) the same data structure types are adopted for the three types of information, and (ii) different data structure types are integrated for storing and retrieving the three different information types. We report our findings in terms of the performance of various query processing tasks such as Boolean and ranked intersection for the different indices and discuss which index type would be appropriate under different conditions for semantic search.     

基于图排序和最大信息增益的领域实体抽取方法

领域知识图谱在各行各业中都发挥着重要作用,领域实体的获取则是构建领域知识图谱的重要基础。数据标注、编写抽取规则等现有的实体抽取方法往往需要较多的人工参与工作。提出一种基于图排序的实体抽取方法和基于最大信息增益的实体扩展方法来构建领域实体集,通过实体识别获得候选实体,基于维基百科的背景信息计算候选实体间的相关度构建实体图,并利用基于置信度传播的图排序算法筛选领域核心实体。在DBpedia中根据最大信息增益来平衡类与领域核心实体相关性及类的抽象程度两个因素以生成实体扩展的共性类。在此基础上,通过SKOS体系中的“Is subject of”关系获得共性类的实例实体,并根据基于字符串相似和结构相关度的方法对扩展实例实体进一步筛选,最终获得全面、准确的领域实体集。以数据结构课程为例构建该课程领域实体集,得到1 115个实体。实验结果表明,在领域数据集上,领域实体抽取F1值达到0.67,能够在较少人工参与的条件下有效获得领域实体,有助于领域知识图谱的构建。     

面向模式图变化的增量图模式匹配

在大数据时代,数据图的规模急剧增长,增量图模式匹配算法能够在数据图或模式图发生变化时避免重新在整个数据图上进行匹配、减少响应时间,因此成为了研究的热点.针对实际应用中数据图不变而模式图发生变化的情况,提出了一种面向模式图变化的增量图模式匹配算法PGC_IncGPM,在模式图匹配的过程中记录适当的中间结果作为索引,用于后续的模式匹配.提出了增强的图模式匹配算法GPMS,用于首次整个数据图上的模式匹配.该算法一方面能够建立后续增量匹配所需的索引,另一方面减少了整个数据图匹配的执行时间.设计实现了面向模式图增边和减边的两个核心子算法,通过子算法的组合,能够支持在模式图发生各种变化时进行增量图模式匹配.在真实数据集和合成数据集上进行实验,结果表明:与重新在整个数据图上进行匹配的ReComputing算法相比,当模式图中变化的边的数目不超过不变的边的数目时,PGC_IncGPM算法能够有效减少图模式匹配的执行时间;随着数据图规模的增大,PGC_IncGPM算法相对于ReComputing算法的执行时间的减少程度更加明显,对于大规模数据图具有更好的适用性.     

基于蚁群优化算法的碎纸拼接

对于碎纸拼接问题,找到一个全局一致的最终解决方案是非常关键的。本文提出了一种基于蚁群优化算法(ACO)的全局拼接方法。首先运用基于ψ-s分析的局部匹配方法进行局部拼接,然后利用局部拼接产生的候选匹配对构建搜索图、信息素与节点相关联。在迭代过程中,利用候选匹配对之间的矛盾和由蚁群构建的全局拼接路径更新信息素:对于在迄今最优路径上的候选匹配对,人工蚂蚁释放信息素;而位于构建失败的路径上的候选匹配对的信息素以一定比例蒸发。候选匹配对的信息素最终向一定方向收敛。最后,根据信息素来筛选候选匹配对。实验证明了这种方法的有效性。     

一种基于图模型的维基概念相似度计算方法及其在实体链接系统中的应用

实体链接是指将文本中具有歧义的实体指称项链接到知识库中相应实体的过程。该文首先对实体链接系统进行了分析,指出实体链接系统中的核心问题—实体指称项文本与候选实体之间的语义相似度计算。接着提出了一种基于图模型的维基概念相似度计算方法,并将该相似度计算方法应用在实体指称项文本与候选实体语义相似度的计算中。在此基础上,设计了一个基于排序学习算法框架的实体链接系统。实验结果表明,相比于传统的计算方法,新的相似度计算方法可以更加有效地捕捉实体指称项文本与候选实体间的语义相似度。同时,融入了多种特征的实体链接系统在性能上获得了达到state-of-art的水平。     

迭代的图变换匹配算法

目的：图像的精确匹配在图像处理与识别中起着重要的作用。为了提高图像的匹配效果,本文提出了一种迭代的图变换匹配算法来实现误匹配关系的去除从而提高图像的匹配精度。方法：该算法首先利用传统的图变换匹配(GTM)算法从初始匹配关系集合中获得较为精确的匹配关系子集,然后,利用已经获得的正确匹配点集与初始匹配点集之间的几何关系对初始匹配进行修正。最后,利用GTM对修正后的匹配关系进一步优化,从而得到更多的精确匹配关系。结果：实验结果显示在不同的图像变换场景下,相比于传统GTM算法,该算法具有较高的查全率。结论：所提算法能够克服传统GTM算法所得正确匹配关系少的缺陷。     

融合实体类型信息的本体-实例联合学习方法

对表示知识图谱的本体图和实例图进行联合学习能够提高嵌入学习效率,但不能区别表示实体在不同场景下的不同意义。在嵌入时考虑三元组中实体的关系类型特征,提出一种融合实体类型信息的本体-实例联合学习方法JOIE-TKRL-CT,达到在联合学习中表示多义实体、提高知识图谱嵌入学习效率的目的。在视图内部关系表示上,利用实体分层类型模型融入实体类型信息,在两个独立的嵌入空间中分别表征学习;在视图间关系表示上,将表征在两个独立空间的本体和实例通过非线性映射的方法跨视图链接。基于YAGO26K-906和DB111K-174数据集的实验结果表明,JOIE-TKRL-CT能够准确捕获知识图谱的实体类型信息,提高联合学习模型性能,与TransE、HolE、DisMult等基线模型相比,其在实例三元组补全和实体分类任务上均获得最优性能,具有较好的知识学习效果。     

Structure and attribute index for approximate graph matching in large graphs

The increasing popularity of graph data in various domains has lead to a renewed interest in developing efficient graph matching techniques, especially for processing large graphs. In this paper, we study the problem of approximate graph matching in a large attributed graph. Given a large attributed graph and a query graph, we compute a subgraph of the large graph that best matches the query graph. We propose a novel structure-aware and attribute-aware index to process approximate graph matching in a large attributed graph. We first construct an index on the similarity of the attributed graph, by partitioning the large search space into smaller subgraphs based on structure similarity and attribute similarity. Then, we construct a connectivity-based index to give a concise representation of inter-partition connections. We use the index to find a set of best matching paths. From these best matching paths, we compute the best matching answer graph using a greedy algorithm. Experimental results on real datasets demonstrate the efficiency of both index construction and query processing. We also show that our approach attains high-quality query answers.     

Structural graph matching using the EM algorithm and singular valuedecomposition

This paper describes an efficient algorithm for inexact graph matching. The method is purely structural, that is, it uses only the edge or connectivity structure of the graph and does not draw on node or edge attributes. We make two contributions: 1) commencing from a probability distribution for matching errors, we show how the problem of graph matching can be posed as maximum-likelihood estimation using the apparatus of the EM algorithm; and 2) we cast the recovery of correspondence matches between the graph nodes in a matrix framework. This allows one to efficiently recover correspondence matches using the singular value decomposition. We experiment with the method on both real-world and synthetic data. Here, we demonstrate that the method offers comparable performance to more computationally demanding methods     

Visual instance mining from the graph perspective

In this paper, we address the problem of visual instance mining, which is to automatically discover frequently appearing visual instances from a large collection of images. We propose a scalable mining method by leveraging the graph structure with images as vertices. Different from most existing approaches that focus on either instance-level similarities or image-level context properties, our method captures both information. In the proposed framework, the instance-level information is integrated during the construction of a sparse instance graph based on the similarity between augmented local features, while the image-level context is explored with a greedy breadth-first search algorithm to discover clusters of visual instances from the graph. This framework can tackle the challenges brought by small visual instances, diverse intra-class variations, as well as noise in large-scale image databases. To further improve the robustness, we integrate two techniques into the basic framework. First, to better cope with the increasing noise of large databases, weak geometric consistency is adopted to efficiently combine the geometric information of local matches into the construction of the instance graph. Second, we propose the layout embedding algorithm, which leverages the algorithm originally designed for graph visualization to fully explore the image database structure. The proposed method was evaluated on four annotated data sets with different characteristics, and experimental results showed the superiority over state-of-the-art algorithms on all data sets. We also applied our framework on a one-million Flickr data set and proved its scalability.     

基于多维信息融合的知识库问答实体链接

知识库问答实体链接任务需要将问句内容精准链接到知识库中实体.当前方法大多难以兼顾链接实体的召回率和精确率,并且仅能根据文本信息对实体进行区分筛选.因此,文中在合并子步骤的基础上,提出融合多维度特征的知识库问答实体链接模型(MDIIEL).通过表示学习方法,将文本符号、实体和问句类型、实体在知识库中语义结构表达等信息整合并引至实体链接任务中,加强对相似实体的区分,在提高准确率的同时降低候选集的大小.实验表明,MDIIEL模型在实体链接任务性能上具有整体性提升,在大部分指标上取得较优的链接结果.