Technology opportunity discovery is a key factor in technological innovation which is closely related to the development of a country and has a great impact on promoting social progress. To find the unexplored areas of technology innovation and present the specific direction of technology development trends, a comprehensive systematic approach group is proposed to analyze patent text data. It includes three steps. (i) patent text data is collected and preprocessed by natural language process methods; (ii) text mining is executed by using supervised machine learning methods, and it is clustered with the k-mean++ algorithm after text representation as keyword vectors; (iii) patents data is visualized in two-dimensional space by generative topographic mapping, and the principal components analysis is used to transform complex multi-dimensional keyword vectors in two-dimensional space. The technology opportunity discovery of proton exchange membrane fuel cells is analyzed by using the proposed comprehensive approaches in the present study. Finally, the patent vacancy is interpreted by its inverse mapping onto the original keyword vector and approved by experts efficiently with high, objectivity and reliability. Furthermore, the present study can be used as a universal methodology to help enterprises and researchers to identify future technology innovations.