A framework for understanding Latent Semantic Indexing (LSI) performance

In this paper we present a theoretical model for understanding the performance of Latent Semantic Indexing (LSI) search and retrieval application. Many models for understanding LSI have been proposed. Ours is the first to study the values produced by LSI in the term by dimension vectors. The framework presented here is based on term co-occurrence data. We show a strong correlation between second-order term co-occurrence and the values produced by the Singular Value Decomposition (SVD) algorithm that forms the foundation for LSI. We also present a mathematical proof that the SVD algorithm encapsulates term co-occurrence information.     

基于粒计算的本体学习研究

近年来,本体学习技术逐渐成为计算机科学领域的一个研究热点,其重要性已在许多方面表现出来,如知识工程、数据库设计和集成、信息检索与获取、软件工程、自然语言处理等,尤其是本体论在Web上的应用导致了语义Web的诞生。本文的主要内容包括:本体论的起源和发展,粒计算研究的主要动机,粒计算的国内外研究现状,本体学习研究现状。     

Return to Babel: Emergent Diversity, Digital Resources, and Local Knowledge

Information system research that recognizes, supports, and enables access of diverse knowledge communities online has become a major concern. Of particular concern is how these digital resources can support and preserve local identities and promote grass-roots involvement in creating the infrastructure of devolved governance. The aim of this article is to expose some of the conceptual obstacles to programs of devolved and local knowledge resources, and to provide an account of agency that recognizes that online communities, and community identity, are essential for eliciting, managing, and sharing local knowledges. Further, this article demonstrates how the implications of these issues impact institutions for managing knowledge that may be extended to all sorts of knowledge communities.     

基于平行语料库的公示语翻译个案调查

公示语是国际化都市、国际旅游目的地语言环境、人文环境的重要组成部分。公示语翻译指的是城市中的标识、指示牌、路牌、标语、单位名称、通告等的外文翻译问题。公示语翻译的内容涉及常用公示语汇、公共设施标示语、交通运输、观光游览等。公示语具有指示性、提示性、限制性、强制性等应用功能。文章就公示语中常见词语"禁止"的翻译,利用"全国公示语语料库"的资源平台对该词的翻译作个案研究,对基于语料库的调查结果进行观察、分析,从语义韵和词语搭配两方面进行讨论,以期对该词在公示语翻译中的使用提供一个合理的依据。     

Social network innovation in the Internet’s global coffee houses: designing a mobile Help Seeking tool in Learning Layers

In this paper, we argue that there is much that we can learn from the past as we explore the issues raised when designing innovative social media and mobile technologies for learning. Like the social networking that took place in coffee houses in the 1600s, the Internet-enabled social networks of today stand accused of being the so-called “weapons of mass distraction” or worse. However, we point out that modern fears about the dangers of social networking are overdone. The paper goes on to present some of the 1930s ideas of Vygotsky. Part of the Learning Layers project builds on this work; we report on extensive initial co-design work and significant barriers with respect to the design of a mobile Help Seeking tool for the health care sector (UK). We then provide an account of how the Help-Seeking tool is being linked to a Social Semantic Server and briefly report on a follow-up empirical co-design study. We conclude by highlighting associated challenges.     

Pronounce differently,mean differently: A multi-tagging-scheme learning method for Chinese NER integrated with lexicon and phonetic features

Named Entity Recognition (NER) aims to automatically extract specific entities from the unstructured text. Compared with performing NER in English, Chinese NER is more challenging in recognizing entity boundaries because there are no explicit delimiters between Chinese characters. However, most previous researches focused on the semantic information of the Chinese language on the character level but ignored the importance of the phonetic characteristics. To address these issues, we integrated phonetic features of Chinese characters with the lexicon information to help disambiguate the entity boundary recognition by fully exploring the potential of Chinese as a pictophonetic language. In addition, a novel multi-tagging-scheme learning method was proposed, based on the multi-task learning paradigm, to alleviate the data sparsity and error propagation problems that occurred in the previous tagging schemes, by separately annotating the segmentation information of entities and their corresponding entity types. Extensive experiments performed on four Chinese NER benchmark datasets: OntoNotes4.0, MSRA, Resume, and Weibo, show that our proposed method consistently outperforms the existing state-of-the-art baseline models. The ablation experiments further demonstrated that the introduction of the phonetic feature and the multi-tagging-scheme has a significant positive effect on the improvement of the Chinese NER task.     

语义相似度Web服务合成策略的应用研究——以物流信息平台为例

研究语义Web服务匹配问题,提高匹配效率。传统Web服务基于关键字匹配,灵活性差,查准率和查全率偏低;语义Web服务引入机器可理解的语义信息,提高了信息共享程度及服务检索成功率,但服务响应慢,系统开销大。为了提高Web服务匹配效率,降低系统成本,提出一种基于语义相似度的Web服务混合匹配策略。首先对语义Web服务匹配问题进行分析,然后建立基于语义相似度的Web服务混合匹配模型,最后在西江物流平台中设计了一个基于语义Web服务的混合匹配框架。实际应用效果表明,混合匹配策略实现了语义Web的更好匹配,为客户提供更为快捷、准确的“寻车配货”服务。     

稀缺性理论与用户认知下的信息检索研究

将稀缺理论与用户认知理论结合,可以重新定义用户在信息检索过程中的认知行为,合理地阐明用户认知行为的发展趋势,提升信息检索研究的有效性与科学性。本文研究分析了稀缺心理对用户认知的动态影响过程,并在稀缺心理的基础上解释了用户潜在信息需求的深层原因。实验结果证明了用户的社交网络行为与信息检索行为之间存在一定的语义关系,通过提取用户社交网络数据,为研究用户个性化需求提供理论依据和参考依据,有利于信息检索个性化服务的实现。     

基于语义加权的引文网络社区划分研究

为提高引文网络社区划分的准确性,以文档之间的语义关系以及引文之间的引用关系为基础,结合词汇在文档中的位置关系等信息,构建基于词汇语义加权的引文网络。通过GloVe模型对词汇向量化以充分利用词汇语义信息,结合WMD模型度量文献之间的相似度,把文档相似度的计算转变为在约束条件下求线性规划最优解的问题,结合文本的内容及结构特征对网络中的边进行赋权,以Louvain社区发现算法对加权后的引文网络进行社区划分,并对划分后的社区进行分析与检验,实验证明GloVe-WMD模型可提高引文网络社区划分的准确度。     

基于多属性规则的生物医学语义关系研究

生物医学领域文献指数级增长亟需快速识别出领域内核心且关键的语义关系,开展领域知识发现研究。本文借鉴DisGeNET基于规则的关系得分方法设计思想,在专家咨询基础上,提出生物医学领域语义关系具有可靠性、重要性和新颖性3种属性,设定每种属性对应的指标及定量计算方法。通过分析结直肠癌领域关系计算结果,验证本文所提关系得分方法在关系重要性排序方面的有效性。该方法考虑维度更加全面,为生物医学领域知识图谱构建、知识发现等提供有益参考。