This dissertation is concerned with the application of materials discovery framework developed in our group to discover high-refractive-index polymers. Development and application of the framework includes four key parts.

In the first part, we present a method to accurately predict the refractive index (RI) of polymers using a combination of first-principles and data modeling. We validated the model with experimental RI values of polymers (Chapter 2). We further benchmark our results using different model chemistries to optimize the tradeoff between the accuracy and computation time (Chapter 3).

The second part covers the development of a molecular library generator (ChemLG) and a virtual high-throughput screening ( ChemHTPS) infrastructure. We demonstrate the applicability of these software suites by providing examples (Chapter 4).

In the third part, we apply ChemLG and ChemHTPS to generate a library of polyimides and compute their RI values, respectively. Using the data generated in this work, we identify structure-property relationships via hypergeometric distribution analysis (Chapter 5).

Finally, we present the application of machine learning to accelerate the process of property prediction. We construct efficient machine learning models to accurately predict the packing density, polarizability, and RI values of organic molecules and characterize them on a massive scale (Chapter 6).