机械设计领域的命名实体识别研究-计算机工程与应用.pdfVIP

100 2017 ，53（20 ） Computer Engineering and Applications 计算机工程与应用 机械设计领域的命名实体识别研究 用 1，2 1，2 1，2 1，2 陈秋瑗 ，程 光 ，李 迪 ，张 建 应 CHEN Qiuyuan1，2 , CHENG Guang1，2 , LI Di1，2 , ZHANG Jian1，2 与 1.北京联合大学 机器人学院，北京 100020 程 2.北京市智能机械创新设计服务工程技术研究中心，北京 100020 g r 1.College of Robotics, Beijing Union University, Beijing 100020, China 工 o 2.Beijing Engineering Research Center of Smart Mechanical Innovation Design Service, Beijing 100020, China . 机 j a CHEN Qiuyuan, CHENG Guang, LI Di, et al. Named entity recognition for mechanical design and manufacturing 算 e c area. Computer Engineering and Applications, 2017, 53 （20 ）：100-104. 计 w. Abstract ：Named entity recognition has wide applications in the area of natural language processing, but the common w methods cannot accurately identify the proper nouns in manufacture area. In order to solve the named entity recognition for mechanical design and manufacturing area, this paper proposes a new machine learning based method. It carefully w finds some statistical features, and then uses the logistic regression algorithm to calculate the tightness between two adja- cent strings. The proposed method can recognize proper nouns more accurately and efficiently for mechanical design and manufacturing area. Key words ：named entity recognition; mechanical design; logistic regression; tightness 摘 要：命名实体识别技术在自然语言处理技术中占有重要的地位，通用的方法不能很好地解决机械领域的识别问 题。基于字符串之间紧密相邻程度等统计特征，定义不同词之间紧密相连的程度，从而识别机械领域的领域词。通 过计算特征值，用逻辑回归的方法确定相邻字串的紧密相邻程度