基于压铸模本体的领域概念自动抽取方法研究

Study on automatic domain concept extraction based on casting mould ontology

概念是本体的核心,人工抽取领域本体概念存在工作量大、速度慢、维护及更新困难等问题。以压铸模领域概念抽取为例,通过分析领域概念分布特点,结合中文分词技术在自然语言处理上的应用,考虑领域概念相关性,提出了一种基于概念相关性的本体概念抽取方法。选取部分压铸模领域文本作为实验样本,利用ICTCLAS软件分词,接着合成词语并进行相关性判断,经领域专家验证得到压铸模领域概念。实验结果表明,该方法提高了压铸模领域概念抽取的效率和准确度,为领域本体高效构建提供了理论和实践基础。

Concepts are vital core of ontology.There will be many problems such as heavy workload,slow speed,difficult for maintenance when concepts are extracted by manual operation.Taking die casting mould ontology for example,an extracting method of ontology concept based on conceptual relativity is put forward by analyzing the distributing characteristics of die casting mould concepts with application of Chinese word segmentation in natural language processing system.And then taking conceptual relativity for reference by choosing parts of die casting mould texts as sample,words are segmented in it firstly by ICTCLAS software,meanwhile words are synthesized and the relativity is evaluated to get die casting mould concepts after verifying through experts.The experimental results show that this method can improve the efficiency and accuracy when extracting die casting mould concepts,which research provides a theoretical and practical basis for building domain ontology efficiently.