Patent big data analysis is to analyze patent document data by big data technologies. A patent document contains diverse and complete information of researched and developed technology, such as patent title and abstract, patent number, issued date, claim, drawing, etc. The velocity of increasing the number of applied patents is so fast. So, patent data have characteristic of big data, 3V (volume, variety, and velocity). In this paper, we propose a graph model for patent analysis because visualization by graph theory is representative method for big data. We can apply the results of patent analysis to R&D planning and technological innovation for company and nation.