A Study on the Efficient R&D Theme Selection Method with Machine Learning

This paper proposes an R&D theme selection method. There are various methods for the theme selection such as the patent analysis and the delphi investigation. The patents and the peer reviewed papers are frequently used as material for the theme selection. Generally, there are three phases for the R&D term selection such as the short-term R&D theme selection, the long-term R&D theme selection, and the medium-term R&D theme selection. The medium-term R&D theme selection is often aimed implementation within 5 years such as an exploratory technology theme. Since it relies on the heuristics knowledge with the technology trends, an efficient selection method is required among the business field. In this paper, we propose a method of selecting the R&D theme using combination of link mining and machine learning based on the public information. As a result, we satisfy predicting technology structure of 5 years later.