Many processes that occur in nature feature complex reaction networks, such as natural oils and resins, and metabolic oxidation reactions. The challenge of studying such system originates in the heterogeneity of the reaction pathways occurring in the material as well as in the growth of the molecules to intractable sizes. In this thesis, a multi-scale modeling methodology based on formal language approach is proposed to describe the chemical system from atomistic level to the level of large polymers molecules. The methodology consists of automatic reconstruction of a finite reaction network describing polymerizing system, solution of the kinetic model providing concentration profiles of various functional groups and crosslinks, and extraction of global properties of a polymer network using findings from random graph theory. Chapter 2 offers a thorough description of the modeling methodology and its modification to include polymer structures. Chapter 3 offers a comprehensive survey on chemistry of linseed oil, which is a bio-based polymer used as an application of our methodology. In Chapter 4, modeling methodology is used to uncover the autoxidation reaction scheme of triolein and to interpret ESI-MS measurements of aged triolein samples. In Chapter 5 the methodology is applied to reconstruct and solve the kinetic model describing drying and degradation of ethyl linoleate (EL), as a model system of linseed oil. Using this application, we draw a parallel between chemistry and language using Chmosky’s idea of formal grammar.