Analytics-as-a-Service (AaaS) has become a more popular discipline in recent years, and many researchers are concerning about how to collect, collate, store, and analyze big data. The aim of this study is to apply the theory of high-level fuzzy Petri nets (HLFPN) to big data analytics platform. The platform features the following advantages: 1) it enables to describe analytical contents through natural language approaches; 2) it can be used to verify analytical processes through modular approaches; 3) it enables to promote fuzzy theory and solving problems through nonlinear equations; 4) it can be employed to generate Map/Reduce programs automatically through the system; 5) it can be used for parallelization, thereby shortening analysis time; and 6) it enables to inquire results through an interface. Finally, we describe the experiments conducted to verify the functions of the platform.