With the intensification of technological innovation competition and the surge in the number of patents in modern society, companies' technology activities such as technology R&D, technology licensing and technology transfer now regularly involve large patent portfolios, as well as a rising risk of patent infringement. However, the existing patent infringement risk assessment methods are only designed for individual patents, which is not enough to meet the companies' practical need for assessing and managing the infringement risk of large patent portfolios. Therefore, this study proposes a combined approach using network analysis and intuitionistic fuzzy technique for order preference by similarity to ideal solution (IF-TOPSIS), which can assess and rank the infringement risk of a company's patent portfolio relative to each of its competitors' patent portfolios, thereby providing a scientific base for the company's management decisions on patent portfolios. In this approach, one-mode weighted patent proximity networks are constructed to extract topology information of patent portfolios, and the IF-TOPSIS method is combined to obtain and rank the risk results under a proper fuzzy framework. The case study uses patent portfolios containing a total of 6736 standard-essential patents (SEPs) in the information and communications technology (ICT) industry as test data, and verifies the practicality and validity of this intelligent approach. Managerial implications for technology managers and patent attorneys are elaborated.