This paper constructs a synonym network using the vector representations of 53,946 words listed in the List of Commonly Used Modern Chinese Words, with words as nodes and synonym relations as edges, resulting in a network with 26,485 nodes and 71,964 edges. The findings indicate that the network exhibits scale-free, small-world properties. Through community detection, the network can be decomposed into several sub-networks, each representing a distinct synonymous semantic field. Consequently, this study may shed light on the complexity of lexicon networks and the modernization of dictionary compilation.