Document Summarization and Information Extraction for Generation of Presentation Slides
Document Summarization and Information Extraction for Generation of Presentation Slides
复制标题
用于生成演示幻灯片的文档摘要和信息提取
DOI：
10.1109/artcom.2009.74
复制
发表时间：
2009
期刊：
影响因子：
--
通讯作者：
T. Geetha
中科院分区：
文献类型：
--
作者：
Harish Mathivanan;M. Jayaprakasam;K. Prasad;T. Geetha
本文提出了一种半自动化的英文文本幻灯片生成技术。本文所讨论的技术被认为是自然语言处理领域的一次开拓性尝试。该技术包括一个信息提取器和一个幻灯片生成器，它结合了某些NLP方法，如分割，分块，摘要等。利用文本的某些特殊语言特征，如词的本体、所发现的名词短语、语义联系、句子中心性等，为了帮助语言处理任务，可以利用两个工具，即有助于分块的MontyLingua和有助于为表示为OWL（本体网络语言）文件的输入文本创建本体的Doddle。该技术的过程包括提取文本、创建本体、识别项目符号的重要短语和生成幻灯片。
In this paper, a semi automated technique to generate slide presentations from english text documents is proposed. The technique discussed in this paper is considered to be a pioneering attempt in the field of NLP (Natural Language Processing). The technique involves an information extractor and a slide generator, which combines certain NLP methods such as segmentation, chunking, summarization etc.., with certain special linguistic features of the text such as the ontology of the words, noun phrases found, semantic links, sentence centrality etc., In order to aid the language processing task, two tools can be utilized namely, MontyLingua which helps in chunking and Doddle helps in creating an ontology for the input text represented as an OWL (Ontology Web Language) file. The process of the technique comprises of extracting text, creating an ontology, identifying important phrases for bullets and generating slides.