基于向量空间的专利设计知识创新性评估算法

摘要

In order to make use of the design knowledge to assist the innovation design, by analyzing the characteristics of the patent design knowledge, this paper uses the verb noun phrase in the patent document to carry on the innovation level feature extraction.By introducing the principle of cosine similarity of vector space, a knowledge innovation evaluation algorithm is designed proposed based on vector space.Taking the innovative design of self locking function as an example, the validity and practicability of the proposed method are verified.%为利用专利中的设计知识辅助创新设计,通过分析专利设计知识的特点,以专利文献中的动名词短语进行创新层次特征抽取.通过引入向量空间的余弦相似度原理,设计了基于向量空间的专利设计知识创新性评估算法,实现专利设计知识的量化评估辅助设计者创新设计.以自锁功能创新设计为例,验证了提出的专利创新性评估方法的有效性以及实用性.