Skip to Content
You are currently on the new version of our website. Access the old version .
AxiomsAxioms
  • Article
  • Open Access

8 December 2022

16 Pages

Technology Opportunity Analysis Based on Machine Learning

,
and
1
Institute of Machine Learning and Big Data, Korea University, Seoul 02841, Republic of Korea
2
Department of Big Data Statistics, Cheongju University, Cheongju 28503, Republic of Korea
3
Institute of Engineering Research, Korea University, Seoul 02841, Republic of Korea
*
Authors to whom correspondence should be addressed.
This article belongs to the Special Issue Statistical Methods and Applications

Abstract

The sustainable growth of a company requires a differentiated research and development strategy through the discovery of technology opportunities. However, previous studies fell short of the need for utilizing outlier keywords, based on approaches from various perspectives, to discover technology opportunities. In this study, a technology opportunity discovery method utilizing outlier keywords is proposed. First, the collected patent data are divided into several subsets, and outlier keywords are derived using the W2V and LOF. The derived keywords are clustered through the K-means algorithm. Finally, the similarity between the clusters is evaluated to determine the cluster with the most similarity as a potential technology. In this study, 5679 cases of unmanned aerial vehicle (UAV) patent data were utilized, from which three technology opportunities were derived: UAV defense technology, UAV charging station technology, and UAV measurement precision improvement technology. The proposed method will contribute to discovering differentiated technology fields in advance using technologies with semantic differences and outlier keywords, in which the meaning of words is considered through W2V application.

Article Metrics

Citations

Article Access Statistics

Multiple requests from the same IP address are counted as one view.