Using Machine Learning to
Predict Startups’ Industry Category
邀 请 人
Entrepreneurship is vital to economics for its role in stimulating innovation and increase employment. However, the data to study China’s entrepreneurship activities is very limited for economists. Registration data from State Administration for Industry and Commerce provide an opportunity to fill this gap. But the data do not cover the startups’ industry category. In this study, using the business description of each startup, we applied Support Vector Machine and K-Means Clustering to its industry category. Our methods perform well on the China Annual Survey of Industrial Firms and the precision is 92%. In this talk, I will also present several other empirical results based on this data.
Dr. Dongbo Shi is an assistance Professor in the School of International and Public Affairs of Shanghai Jiao Tong University. Prof. Shi’s research focus on applying data science techniques to study issues of science and innovation. He got his PhD degree in public policy from Tsinghua University.