Search results

This paper seeks to propose a method of discovering uncommercialized research fronts by comparing scientific papers and patents. A comparative study was performed to measure the semantic similarity between academic papers and patents in order to discover research fronts that do not correspond to any patents.

The authors compared structures of citation networks of scientific publications with those of patents by citation analysis and measured the similarity between sets of academic papers and sets of patents by natural language processing. After the documents (papers/patents) in each layer were categorized by a citation‐based method, the authors compared three semantic similarity measurements between a set of academic papers and a set of patents: Jaccard coefficient, cosine similarity of term frequency‐inverse document frequency (tfidf) vector, and cosine similarity of log‐tfidf vector. A case study was performed in solar cells.

As a result, the cosine similarity of tfidf was found to be the best way of discovering corresponding relationships.

This proposed approach makes it possible to obtain candidates of unexplored research fronts, where academic researches exist but patents do not. This methodology can be immediately applied to support the decision making of R&D investment by both R&D managers in companies and policy makers in government.

This paper enables comparison of scientific outcomes and patents in more detail by citation analysis and natural language processing than previous studies which just count the direct linkage from patents to papers.