| 赵蕴华.基于深度标引的专利文本挖掘框架研究[J].数字图书馆论坛,2008,(11): |
| 基于深度标引的专利文本挖掘框架研究 |
| The Study of text mining framework for patent document based on deep content indexing |
| 投稿时间：2008-10-07 修订日期：2008-10-07 |
| DOI： |
| 中文关键词: 专利分析，深度标引，文本挖掘，知识抽取 |
| 英文关键词: Patent analysis, Deep content indexing, Text mining |
| 基金项目:其它 |
|
| 摘要点击次数: 1380 |
| 全文下载次数: 661 |
| 中文摘要: |
| 专利文献中的文摘、权利要求项、全文等文本信息蕴涵了重要技术细节和技术保护等内容，从这些专利文本内容中挖掘具有技术、商业价值的潜在信息是当前专利信息应用领域的研究热点。本文研究将面向分析目标的专利文本深度标引应用到专利文本挖掘中，在数据预处理阶段就将分析目标作为知识抽取的基础，专利分析人员可依据分析需求，在文本挖掘时只提取标引结果的某一部分进行分析和处理，这不仅可提高专利文本挖掘的数据预处理质量，也可提高后期文本分析的效率。 |
| 英文摘要: |
| Important technical details and protected contents of patent are included in the abstract, claims and full text. Mining the technical and commercial information from patent contents is taken attention in the research field of patent analysis. In this paper, deep content indexing orient to patent analysis is introduced into the text mining framework for patent documents. In the state of data pretreatment, analysis subject is considered into the process of text refine. Patent analyst can select one or more indexing parts for text mining which can improve the quality of data pretreatment and the efficiency of patent text analysis. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
