论文部分内容阅读
结合理论和实验比较分析用于词形规范的词形还原方法和工具。归纳现有词形还原方法的主要分类,分析各类方法的特点和不足。介绍7种词形还原实现工具,并从其实现原理、使用的词性标注器、词典、开发语言、处理的语种、是否具有拼写检查功能等方面比较分析各工具的特点。选取其中5种工具,利用WordSimith Tools的标准数据进行词形还原实验。结合实验结果分析各工具的优劣,发现Specialist NLP Tools的词形还原工具具有较好的词形还原处理效果,为研究者选择适当的词形还原方法和工具提供参考。
Combining theory and experiment to compare and analyze the inflection method and tool used in inflection of inflexion. Summarize the main classification of the existing form reduction methods and analyze the characteristics and shortcomings of various kinds of methods. This paper introduces seven kinds of morphing implementation tools, and analyzes the features of each tool from the aspects of its realization principle, part-of-speech tagging device, dictionaries, development languages, language processing, and spelling checking function. Select five kinds of tools, the use of WordSimith Tools standard data for shape reduction experiments. Combined with the experimental results, the advantages and disadvantages of each tool are analyzed. It is found that Specialist NLP Tools’ morphology restoration tool has a good performance of shape restoration and provides a reference for researchers to choose the appropriate method and tool of shape reduction.