Abstract: In some embodiments, a computer-implemented method for selecting a polymer for an intended use. The method may include: training a selection model based on a graphical dataset, the graphical dataset comprising a node-edge graph identifying relationships between uses and properties of a polymer; receiving, via a user interface, a request comprising the intended use; based on the selection model, determining a polymer for the intended use; and outputting the determined polymer via the user interface.

Abstract: In some embodiments, a computer-implemented method for selecting a polymer for an intended use. The method may include: training a selection model based on a graphical dataset, the graphical dataset comprising a node-edge graph identifying relationships between uses and properties of a polymer; receiving, via a user interface, a request comprising the intended use; based on the selection model, determining a polymer for the intended use; and outputting the determined polymer via the user interface.

Abstract: Systems and methods are provided for query and index optimization for retrieving data in instances of a formulation data structure from a database. The methods include presenting an information source for searching for the presence of formulations and generating formulation data from field entries. The formulation data is associated with found formulations. The methods include generating an instance of a formulation data structure. The instance of the formulation data structure associates the information source with the found formulations. The methods include creating optimized index data from retrieved data in the instance of the formulation data structure. The optimized index data includes a mapping between potential search-field terms and the formulation data, and is grouped based on a predicted access pattern.