基于MesH领域本体的多语种医学文献分类
Multilingual Medical Documents Classification Based on MesH Domain Ontology
摘要：
本文讨论语义Web和本体。它解决了多语言Web文档的分类问题，基于领域本体。目标是能够使用模型对不同语言的文件进行分类。我们将尝试使用两种不同的方法来解决这个问题。这两种方法将有两个基本阶段：在标记语料库上使用机器学习算法创建模型，然后在检测其语言并将其术语映射到参考语言（英语）的概念后对文档进行分类。但每一个国家都将以不同的方式处理多语制问题。一种认为本体是单语的，而另一种认为它是多语言的。为了展示我们工作的可行性和重要性，我们在一个领域上实现了它，这个领域现在吸引了数据采掘社区的很多关注：生物医学领域。所选文档来自生物医学基准语料库Ohsumed，相关本体是同义词库MeSH（医学主题词）。我们工作的主要思想是一个新的文档表示，所有好的分类的杰作，基于概念。实验结果表明，所推荐的思路是有前途的。
英文摘要：
This article deals with the semantic Web and ontologies. It addresses the issue of the classification of multilingual Web documents, based on domain ontology. The objective is being able, using a model, to classify documents in different languages. We will try to solve this problematic using two different approaches. The two approaches will have two elementary stages: the creation of the model using machine learning algorithms on a labeled corpus, then the classification of documents after detecting their languages and mapping their terms into the concepts of the language of reference (English). But each one will deal with the multilingualism with a different approach. One supposes the ontology is monolingual, whereas the other considers it multilingual. To show the feasibility and the importance of our work, we implemented it on a domain that attracts nowadays a lot of attention from the data mining community: the biomedical domain. The selected documents are from the biomedical benchmark corpus Ohsumed, and the associated ontology is the thesaurus MeSH (Medical Subject Headings). The main idea in our work is a new document representation, the masterpiece of all good classification, based on concept. The experimental results show that the recommended ideas are promising.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
