Topic discovery based on text mining techniques     

Aurora Pons-Porrata  Rafael Berlanga-Llavori  José Ruiz-Shulcloper 《Information processing & management》2007

In this paper, we present a topic discovery system aimed to reveal the implicit knowledge present in news streams. This knowledge is expressed as a hierarchy of topic/subtopics, where each topic contains the set of documents that are related to it and a summary extracted from these documents. Summaries so built are useful to browse and select topics of interest from the generated hierarchies. Our proposal consists of a new incremental hierarchical clustering algorithm, which combines both partitional and agglomerative approaches, taking the main benefits from them. Finally, a new summarization method based on Testor Theory has been proposed to build the topic summaries. Experimental results in the TDT2 collection demonstrate its usefulness and effectiveness not only as a topic detection system, but also as a classification and summarization tool.  相似文献   

Interest in an expository text: How does it flow from reading to writing?   总被引：1，自引：0，他引：1  

Pietro Boscolo  Nicola AriasiLaura Del Favero  Chiara Ballarin 《Learning and Instruction》2011,21(3):467-480

This study investigated the flow of interest in a reading-to-write activity. It was hypothesized that (a) different contents in a text would stimulate different types of interest and (b) different writing tasks would impact differently students’ use of interesting text segments. Participants were 247 11th- and 12th-grade students, who were divided into 6 groups according to a 2 (interesting vs. uninteresting topic) × 3 (type of writing task) design. While reading, students rated novelty, curiosity, impressiveness, importance, and willingness to reflect on each text segment. Participants were then assigned to one of three writing tasks, namely argumentation, text designing, and composition. Results showed that concepts were regarded as important, events/activities as impressive, and evaluations as most worthy of reflection. Topic interest was explicitly expressed mainly in the argumentation condition, where increased topic interest scores after reading and writing were also found.  相似文献   

基于主题句相似度的标题党新闻鉴别技术研究     

王志超  翁楠  王宇 《现代图书情报技术》2011,(11):48-53

针对目前网络上的标题党新闻,提出一种标题党新闻自动识别的算法。通过分析新闻网页构成的特点,抽取出新闻标题和新闻正文;以句子关系矩阵为基础,提出一种以语句为单位的主题句抽取算法;根据句子相似度计算结果来进行判断。实验表明,本方法的识别精度可达到80%,是一种有效的方法。  相似文献   

基于种子文档LDA话题的演化研究   总被引：1，自引：0，他引：1  

单斌  李芳 《现代图书情报技术》2011,(Z1):104-109

提出一种基于种子文档的LDA话题演化方法。首先选取种子文档,利用种子文档指导后一时间段文档的建模,然后根据种子文档的语义分布信息对连续时间上的LDA话题进行关联,保证话题的同一性。实验结果证明,在NIPS论文语料集和全国两会新闻报道集中,该方法可以推导特定话题的演化结果,避免关联话题之间存在的演化结果。  相似文献   

基于Single-Pass的网络话题在线聚类方法研究     

朱恒民  朱卫未 《现代图书情报技术》2011,(12):52-57

基于Single-Pass算法思想,研究网络话题的在线聚类方法,以期及时捕捉网络信息的动态变化在分析该方法聚类流程的基础上,重点研究网络动态信息流的文本特征抽取和权重计算方法,以及话题类表示和更新等关键问题,设计实验对比分析不同的标题中特征加权系数、特征权重计算和标准化方法以及话题类向量维度对话题聚类质量和时间效率的影响。  相似文献   

科技信息资源内容监测与分析服务平台概况     

徐硕  乔晓东  朱礼军  张运良 《数字图书馆论坛》2011,(11):38-44

为了方便研究者分析领域的研究动态,了解领域内研究的重要研究者和重要文献,并对科技文献和科技工作者的工作进行准确的评价,作者借助国家科技图书文献中心（NSTL）雄厚的资源优势,联合清华大学等有关优势单位,共同开发了面向西文资源的科技信息资源内容监测与分析服务平台,该平台具有专家、期刊/会议和关键词统一检索功能,具有研究者关联路径发现、主题发现等功能,并且内嵌了专家和论文排名功能。  相似文献   

Fifty years of information management research: A conceptual structure analysis using structural topic modeling     

《International Journal of Information Management》2021

Information management is the management of organizational processes, technologies, and people which collectively create, acquire, integrate, organize, process, store, disseminate, access, and dispose of the information. Information management is a vast, multi-disciplinary domain that syndicates various subdomains and perfectly intermingles with other domains. This study aims to provide a comprehensive overview of the information management domain from 1970 to 2019. Drawing upon the methodology from statistical text analysis research, this study summarizes the evolution of knowledge in this domain by examining the publication trends as per authors, institutions, countries, etc. Further, this study proposes a probabilistic generative model based on structural topic modeling to understand and extract the latent themes from the research articles related to information management. Furthermore, this study graphically visualizes the variations in the topic prevalences over the period of 1970 to 2019. The results highlight that the most common themes are data management, knowledge management, environmental management, project management, service management, and mobile and web management. The findings also identify themes such as knowledge management, environmental management, project management, and social communication as academic hotspots for future research.  相似文献   

话题型微博中第三人称代词消解     

张瑛 《海南广播电视大学学报》2014,(1):29-32

对话题型微博中第三人称代词的使用特征进行分析。分析表明,话题型微博人称代词用法不规范;常出现外指;非指称性人称代词较多。针对这些特征,提出了基于多层面语言信息的消解策略,第三人称代词消解实验结果 F指数达到了85.83%。  相似文献   

通过一题多解寻求一题巧解培养创新思维     

王绍恒 《宜宾学院学报》2005,5(12):21-24

本文通过对高等数学中的几个典型例题或习题的解法进行多角度,全方位的探讨,给出多种解法以便寻求出巧妙的或最佳的解法,从而引导学生形成广阔的多方位的解题思路,达到扩展学生思雏的目的.  相似文献   

信息结构中调核的非焦点功能解释     

余玲丽 《玉溪师范学院学报》2005,21(4):89-93

调核是提示信息焦点的一种重要手段。然而,调核有时提示的不是信息焦点,而落在语用预设命题之中。语用预设命题成分是否接受调核取决于两个因素:其所指的激活状态以及与大命题之间的语用关联。决定调核的位置不是由语用预设命题的认知状态所决定的,而是由其命题所指与大命题之间的语用关联性决定的。  相似文献