Case Study with Matmatch
This case study describes the ability of Alchemite™ to populate missing material property data and add significant value to Matmatch’s services. It also shows that Intellegens’ machine learning tools were easily integrated with Matmatch’s systems and that their data scientists could effortlessly build complex models of multi-dimensional data with Alchemite™ software.
Alchemite™ was able to populate sparse and missing data and model complex multi-dimensional data.
Data scientists from Matmatch found it easy to access the full power of Alchemite™ via the Alchemite™ Engine API, which integrated well with their systems.
Alchemite™ provided a significant value and potential for additional revenue to Matmatch through data checking, estimation, and gap filling.
“We were most impressed by the results we were able to achieve using the Intellegens Alchemite tool to improve the Matmatch material data set. The outcome exceeded our expectations and we succeeded in predicting and filling significantly more properties than expected across different material categories despite starting with a very sparse dataset. The Intellegens team was a pleasure to work with and we look forward to continued cooperation.“
– Melissa Albeck, CEO at Matmatch.
Material design is a highly complex endeavor, mainly due to the vast amount of possible compositions and process parameters, but also due to the time and cost associated with experimentally measuring material properties of interest. Selecting a material that achieves the required set of properties (e.g., strength, corrosion resistance, weight, or cost) is essential for the success of a project.
To help select optimal materials, Matmatch provides extensive databases where characterized materials can be browsed, compared, and matched to customer project needs. However, these databases are sparse, as not every material property is available for each material. The goal of this project was to use the Alchemite™ machine learning toolkit to impute (populate) the missing data from the databases as well as to verify the existing data by identifying outliers.