Abstract: Park green space, an important component of the urban green space system, provides natural ecological functions such as rain flood management, air purification, and ecological maintenance, and plays an important role in improving leisure experience, strengthening neighborhood communication, and influencing real estate values in terms of social and economic benefits. However, there are some problems with the methods of evaluating the accessibility of green space in parks, such as the singularity of the perspectives and indexes. The space syntax theory provides an effective tool for quantitative analysis and evaluation of the park layout space researches. Taking 16 comprehensive parks within the Hangzhou Ring Highway as the objects, the research applies the space syntax theory, as well as the kernel density analysis and buffer analysis in the ArcGIS, to analyze the accessibility of comprehensive parks in Hangzhou. It also adopts the LBS big data to study the real use intensity of comprehensive parks, and further verify the result of syntactic measurement from the angle of practical applications. The results show that the overall and local accessibility of the parks is good, while the perceptual accessibility is quite different. The spatial distribution of comprehensive parks and the shapes of the road networks are partly “dislocated”, and the use efficiency of the parks on working days and weekends presents three similar stages of “rising - steady - falling”, the conclusion of population aggregation in LBS big data is slightly different from that in theory in the local scale, though the overall space tends to be consistent. According to the results, it proposes that the overall accessibility of the comprehensive parks can be enhanced by increasing the density of the road network or the number of parks.