【作者】 孙涌； 韦小丽；

【Author】 SUN Yong~1,WEI Xiaoli College of Computer Science and Technology,Soochow University,Jiangsu Suzhou 215006,China;

【机构】 苏州大学计算机科学与技术学院；

【摘要】 本体是语义检索的核心。本体构建主要包括领域概念获取和概念间关系获取,其中领域概念获取是本体构建的基础。本文采用基于最大熵模型的方法来获取概念,该模型首先通过对领域文本进行挖掘而得到名词性短语,然后使用改进的TF-IDF公式从中抽取具有领域性的短语,经人工修正后得到本体概念。本工作以专利服务平台为背景,并以电梯专利信息为例进行实际应用,结果表明此方法提高了概念的准确性和完整性。更多还原

【Abstract】 Ontology is the core of the semantic retrieval.Ontology construction mainly includes concept extraction and the extraction of relationship between concepts,the concept extraction is the base of ontology construction.In this paper,the domain-specific concepts were extracted by the approach which was based on the maximum entropy model,we firstly mined the base noun phrases from the texts in the field,then extracted the domain-specific phrases from the phrases,and finally we corrected the phrases to form the ontology concept.The incensement of the accuracy and completeness of the concepts has been demonstrated by the experimental results.更多还原