Exploring similarity between academic paper and patent based on Latent Semantic Analysis and Vector Space Model
Hongjiao Xu, Wen Hua Zeng, Jie Gui, Peng Qu, Xiaohua Zhu, Lijun Wang · 2015
With the development of network technology, the storage format of science and technology literature changes from paper to electronic version, and its size also is increasing. The academic papers and patents are important science and technology literature. To a certain extent, they represent the highest level of academic research and technical innovation. In this paper, we perform a study to measure the semantic similarity between academic papers and patents. The paper argues it's important to get similarity between single paper and single patent. To find linkage between them, four semantic similarity measurements are compared: Latent Semantic Analysis (LSA) based on words, LSA based on terms, Vector Space Model (VSM) based on words, VSM based on terms. A case study is conducted in the area of optical sensors. And result shows that the measurement method of terms based VSM is the best to find the similarity between single paper and single patent.