Study on association rules mining based on semantic relativity

An association rules mining method based on semantic relativity is proposed to solve the problem that there are more candidate item sets and higher time complexity in traditional association rules mining.Semantic relativity of ontology concepts is used to describe complicated relationships of domains in the method.Candidate item sets with less semantic relativity are filtered to reduce the number of candidate item sets in association rules mining.An ontology hierarchy relationship is regarded as a directed acyclic graph rather than a hierarchy tree in the semantic relativity computation.Not only direct hierarchy relationships,but also non-direct hierarchy relationships and other typical semantic relationships are taken into account.Experimental results show that the proposed method can reduce the number of candidate item sets effectively and improve the efficiency of association rules mining.

Approximate querying between heterogeneous ontologies based on association matrix

An approximate approach of querying between heterogeneous ontology-basedinformation systems based on an association matrix is proposed. First, the association matrix isdefined to describe relations between concepts in two ontologies. Then, a methodof rewriting queriesbased on the association matrix is presented to solve the ontology heterogeneity problem. Itrewrites the queries in one ontology to approximate queries in another ontology based on thesubsumption relations between concepts. The method also uses vectors to represent queries, and thencomputes the vectors with the association matrix; the disjoint relations between concepts can beconsidered by the results. It can get better approximations than the methods currently in use, whichdonot consider disjoint relations. The method can be processed by machines automatically. It issimple to implement and expected to run quite fast.

Context query and association discovery for collaborative environment

A context memory model and an approach for context query and association discovery are proposed. The context query is based on a resource description framework （RDF） dataset and SPARQL language. To discover collaborative associations, an approach of transforming RDF named graphs into ＂context graph＂ is proposed. First, the definitions of the importance of the nodes and the weight assignment for the ＂context graph＂ are given. Secondly, the implementation of a spread activation algorithm based on ＂context graph＂ is proposed. An infrastructure is also built up in the collaborative context space （CCS） system to support context memory and knowledge discovery in a collaborative environment.

Applying Association Approach——A Guide to Effective Vocabulary Teaching

Vocabulary, as the basic units of language, has recently gained the recognition it deserves. However, there are few enough topics on the vocabulary teaching. This paper, based on the traditional methods, puts forward association approach in vocabulary teaching and illustrates its detailed classification and teaching procedure.

Query Through Heterogeneous Ontologies Using Association Matrix

This paper introduces the definition and calculation of the association matrix between ontologies. It uses the association matrix to describe the relations between concepts in different ontologies and uses concept vectors to represent queries; then computes the vectors with the association matrix in order to rewrite queries. This paper proposes a simple method of querying through heterogeneous Ontology using association matrix. This method is based on the correctness of approximate information filtering theory; and it is simple to be implemented and expected to run quite fast. Key words semantic Web - information retrieval - ontology - query - association matrix CLC number TP 391 Foundation item: Supported by the National Natural Science Foundation of China (60373066, 60303024), National Grand Fundamental Research 973 Program of China (2002CB312000) and National Research Foundation for the Doctoral Program of Higher Education of China (20020286004)Biography: KANG Da-zhou (1980-), male, Master candidate, research direction: Semantic Web, knowledge representation on the Web.

STUDY ON SEMANTIC-BASED VIDEO WATERMARKING METHOD

A new video watermarking method for the Audio Video coding Standard (AVS) is proposed. According to human visual masking properties, this method determines the region of interest for watermark embedding by analyzing video semantics, and generates dynamic robust watermark according to video motion semantics, and embeds watermarks in the Intermediate Frequency (IF) Discrete Cosine Transform (DCT) coefficients of the luminance sub-block prediction residual in the region of interest. This method controls watermark embedding strength adaptively by video textures semantics. Ex- periments show that this method is robust not only to various conventional attacks, but also to re-frame, frame cropping, frame deletion and other video-specific attacks.

Document Clustering Using Semantic Cliques Aggregation

The search engines are indispensable tools to find information amidst massive web pages and documents. A good search engine needs to retrieve information not only in a shorter time, but also relevant to the users’ queries. Most search engines provide short time retrieval to user queries;however, they provide a little guarantee of precision even to the highly detailed users’ queries. In such cases, documents clustering centered on the subject and contents might improve search results. This paper presents a novel method of document clustering, which uses semantic clique. First, we extracted the Features from the documents. Later, the associations between frequently co-occurring terms were defined, which were called as semantic cliques. Each connected component in the semantic clique represented a theme. The documents clustered based on the theme, for which we designed an aggregation algorithm. We evaluated the aggregation algorithm effectiveness using four kinds of datasets. The result showed that the semantic clique based document clustering algorithm performed significantly better than traditional clustering algorithms such as Principal Direction Divisive Partitioning (PDDP), k-means, Auto-Class, and Hierarchical Clustering (HAC). We found that the Semantic Clique Aggregation is a potential model to represent association rules in text and could be immensely useful for automatic document clustering.

基于深度学习的移动机器人语义SLAM方法研究

为了给移动机器人提供细节丰富的三维语义地图,支撑机器人的精准定位,本文提出一种结合RGB-D信息与深度学习结果的机器人语义同步定位与建图方法。改进了ORB-SLAM2算法的框架,提出一种可以构建稠密点云地图的视觉同步定位与建图系统;将深度学习的目标检测算法YOLO v5与视觉同步定位与建图系统融合,反映射为三维点云语义标签,结合点云分割完成数据关联和物体模型更新,并用八叉树的地图形式存储地图信息;基于移动机器人平台,在实验室环境下开展移动机器人三维语义同步定位与建图实验,实验结果验证了本文语义同步定位与建图算法的语义信息映射、点云分割与语义信息匹配以及三维语义地图构建的有效性。

LiDar点云指导下特征分布趋同与语义关联的3D目标检测

针对现有基于伪点云的3D目标检测算法精度远低于基于真实激光雷达(Light Detection and ranging,LiDar)点云的3D目标检测,本文研究伪点云重构,并提出适合伪点云的3D目标检测网络.考虑到由图像深度转换得到的伪点云稠密且随深度增大逐渐稀疏,本文提出深度相关伪点云稀疏化方法,在减少后续计算量的同时保留中远距离更多的有效伪点云,实现伪点云重构.本文提出LiDar点云指导下特征分布趋同与语义关联的3D目标检测网络,在网络训练时引入LiDar点云分支来指导伪点云目标特征的生成,使生成的伪点云特征分布趋同于LiDar点云特征分布,从而降低数据源不一致造成的检测性能损失;针对RPN(Region Proposal Network)网络获取的3D候选框内的伪点云间语义关联不足的问题,设计注意力感知模块,在伪点云特征表示中通过注意力机制嵌入点间的语义关联关系,提升3D目标检测精度.在KITTI 3D目标检测数据集上的实验结果表明:现有的3D目标检测网络采用重构后的伪点云,检测精度提升了2.61%;提出的特征分布趋同与语义关联的3D目标检测网络,将基于伪点云的3D目标检测精度再提升0.57%,相比其他优秀的3D目标检测方法在检测精度上也有提升.

知识关联视角下标准文档的多粒度知识组织方法研究

传统的文档组织方式无法应对标准数字化发展形势,有必要充分发掘标准文档中的多粒度知识单元及其语义关联,探索能够高效运用标准知识的新型组织方法,为优化标准供给提供参考。从知识关联视角出发,提出一种面向标准文档的多粒度、富语义的通用知识组织方法。首先,基于知识粒度理论,依据标准文档的知识内容和需求特征进行多粒度的知识划分与描述;其次,从知识层级、文档特征、文本逻辑、时空演化等方面认知和发现标准多粒度知识间的语义关联模式与类型;最后,采用本体构建方法实现标准文档的多粒度知识组织,并通过知识实例的添加来实现本体验证与价值阐述。多粒度知识关联的标准组织方法能够完整揭示标准文档中的多粒度知识单元,形成联通广泛的知识层次与关联,有助于标准知识在多种服务场景中被有效获取、共享与重用,既推进了适应数智时代的标准资源建设,又丰富了多粒度知识驱动的文档内容挖掘与利用。

“路径”和“模式”的内涵辨析

“路径”“模式”作为基本概念,被广泛应用在各学科领域研究中,出现了各类混用、误用现象,并且这些现象仍在蔓延。厘清二者的内涵是一项基础性理论工作。总体上,“路径”是多条的、有范围的,由初始节点、关键节点和终结节点组成,节点之间按照一定次序连接。“模式”具有可复制属性,是相关要素按照一定规则形成的组织形式,且要素之间存在一定的内在关系。但是,“路径”“模式”的内涵在不同学科领域的研究中有不同的表达,尤其是“模式”表现得更加明显。

《文心雕龙》格式塔意象翻译研究——以施友忠英译本为例

《文心雕龙》格式塔意象在构建和传达文学理论方面发挥着重要作用,它们来源于作者认知世界中的意义载体或心理映射,与英文格式塔意象存在着巨大差异。如何再造译文格式塔意象是译者无法推卸的责任。施友忠分别采用等效移用、变异增益、融通新创的翻译策略建构译文格式塔意象,而支撑这些翻译策略的是译者认知、读者认知和混合认知三种翻译路径。三种翻译路径作为文论逻辑建构的网络节点,充当了认知开关的按钮,激发了读者关于意象的语义联想,疏通了读者关于文论思想的解读通道,拓展了译文可理解和接受的空间,也为《文心雕龙》在异域文化语境的传播与接受打下了坚实的基础。

语音邻域密度、词性类别、语言水平对二语词汇联想反应的影响

本文研究语音邻域密度、词性类别及语言水平对中国英语学习者二语词汇联想反应的影响。使用Lex_30测量词汇水平,采用“听—写”词联想任务考察所选词的语音邻域稀疏与密集程度以及名、动、形不同词性的刺激词联想反应差异。结果表明:(1)高低水平学生产出的语音和语义反应均存在显著差异,语音反应比例与语言水平成负向相关;(2)不同词性的语义联想反应有显著差异,但在语音反应上无差异;(3)不同语音邻域密度的刺激词只在高语言水平者的语义反应上有显著差异,低水平者只有名词的语音及语义反应存在显著差异,部分支持邻域激活模型。

标准知识数字化表达通用模型与自动抽取技术研究

标准数字化是国内外标准化发展的重要领域和方向。研究以食品与农产品领域标准为研究对象开展标准知识数字化表达的通用模型与自动抽取技术研究,明确了国内外标准化数字化研究现状与问题,提出了标准知识数字化表达的通用模型,开展了标准知识数字化自动提取技术研究,实现对表达模型的知识要素的自动标注和抽取,并据此形成具有语义关联的标准知识。最后,以2000项食品与农产品领域标准为例进行标准知识数字化表达模型与自动提取技术的实证研究。

联合语义分割的自监督单目深度估计方法

研究深度估计和语义分割的图像之间的互利关系,提出了一种联合语义分割的自监督单目深度估计方法USegDepth.语义分割和深度估计任务通过共享编码器,实现语义引导.为了进一步提高编码器的跨多任务性能,设计了多任务特征提取模块,堆叠该模块构成共享编码器,解决有限感受野和缺乏跨通道交互导致的模型特征表示能力欠佳问题,进一步提升模型精度.同时,提出跨任务交互模块,通过双向的跨域信息交互细化特征表示,提升深度估计表现,特别是光度一致性监督有限的弱纹理区域和物体边界.通过在KITTI数据集上的训练和全面评估,实验结果显示所提的USegDepth模型方法的均方相对误差相比于SGDepth降低了0.176个百分点,在阈值为1.253的阈值精度达到了98.4%,证明了USegDepth在深度预测上具有较高的准确率.

企业技术创新风险的信息聚合及服务体系框架构建研究

借鉴企业技术创新风险相关研究和基于语义的信息聚合模式,利用LDA主题模型和本体建模技术,开展基于科创板企业披露文本的风险主题聚合和基于风险事件案例的语义关联聚合,构建包含用户需求层、数据资源层、关联聚合层和应用服务层的企业技术创新风险信息服务体系框架。企业技术创新风险主题涵盖技术、财务、人力、生产、管理、市场、政策等方面的不确定性,语义关联聚合涉及风险的因素、事件、影响和防控4个概念的信息抽取,开展相关信息服务需要综合考虑“需求驱动”和“资源导向”,提供企业技术创新风险画像和智能决策服务。

历史地图知识组织:需求、框架及实践

历史地图是信息的重要载体之一,其信息隐含在由多种元素组成的复杂网络中。为了高效发挥历史地图的地理空间表示作用,深入挖掘其历史价值、人文价值与证据价值,需要构建统一的历史地图知识组织框架与语义组织模型。本文从形态描述需求、内容描述需求、地理知识服务、文化知识服务、证据知识服务五个维度对历史地图的服务目标进行系统分析,获得“读图者”与“资源管理者”对历史地图的知识组织需求。在此基础上,设计了包含历史地图文本化、概念识别、本体构建以及知识关联路径与关联模式等核心环节的历史地图知识组织框架,并阐述各环节及其之间的核心含义与逻辑关系,形成了历史地图的结构化、数据化、语义化组织机制。本研究为历史地图提供了完整的语义描述范式,丰富了多源异构信息载体的知识组织理论与方法。

多源自媒体资源知识组织模型构建研究

[目的/意义]多源自媒体资源已成为获取信息的重要途径,其多源异构特性使有效知识组织面临挑战。鉴于当前缺乏统一的媒体资源元数据标准,本研究通过文本化、知识元抽取和语义关联等技术路径,旨在构建和实现针对多源自媒体资源的知识组织模型。[方法/过程]研究首先复用和自定义DC元数据,构建多源自媒体资源的元数据描述框架,并据此构建多源自媒体资源本体和语义网络。借助计算机算法实现多源自媒体资源的文本构建标准化,使知识组织研究可以深入到多源自媒体资源的具体内容中。此外,还实现了多源自媒体资源的命名实体、关键词和知识摘要的抽取,并设计了语义关联实验,以上研究都以B站、抖音、知乎等平台数据为基础。[结果/结论]通过算法在多源自媒体资源和给定虚拟馆藏资源进行关联,达成了虚拟馆藏资源到多源自媒体资源的语义关联映射,旨在发掘多源自媒体资源的多元价值,为图书馆的知识服务提供新思路。

面向功能语义增强与标签关联的Web服务标签推荐

为了提升标签推荐的质量,提出一种面向功能语义增强与标签关联的Web服务标签推荐方法。将语境权重融入TextRank模型,提取与服务功能契合度高的关键词,用于构建功能语义增强的服务表征向量;建立标签关联图,基于改进的GraphSAGE模型生成标签关联向量;利用KNN算法获取推荐的主标签与候选标签集合,面向服务表征向量和标签关联向量构建融合适配度与关联度的标签推荐方法。实验表明,所提方法在accuracy与F_(1)-score指标上优于当前流行的标签推荐方法,标签推荐质量得到提升。

基于知识元迁移的学科领域方法库构建研究

[目的/意义]学科领域方法库构建是在特定学科领域中收集、存储相关方法和技术的过程,它能够为解决领域问题提供有效参考。通过知识元迁移整合跨学科的研究方法,丰富和更新本学科的方法库,对本领域学术研究和实践创新具有指导意义。[方法/过程]以计算机科学的方法向情报学迁移为例,运用规则抽取获得两学科的方法知识元,在此基础上通过分析学科间方法的逻辑关系和相似关系实现方法知识元迁移。[结果/结论]研究表明,基于逻辑关系迁移的方法知识元有:FSDP聚类算法、DE-LSTM预测方法、TF-LDA模型等。基于相似关系迁移的方法知识元有:FastICA算法、KM匹配、蒙特卡罗算法、Q-学习、帧差法等。本研究不仅扩展了情报学方法库,为情报学领域的问题解决提供新的思路,同时也为学科间的知识交流和交叉学科研究提供了有价值的参考。