Evaluation of Descriptors for Archetypal Analysis of Metallic Nanoparticles



Metallic nanoparticles are used as in a range of biomedical and energy applications, and are the focus of intense research. The structure of the individual nanoparticles (only millions of a millimetre in size) determines the medical efficacy or energy efficiency, and manufacturers are keen to understand which aspects of the nanoparticle structure could be used to target different applications. Special cases such as the archetypes can be used to guide experimentation and provide a target for synthesis. However, there are many ways to describe the structure of a nanoparticle, and it is unclear at this stage if the final archetypes are sensitive this issue.  While some work has been done using archetypal analysis to study the pure types of metallic nanoparticles, a detailed study of the impact of the choice of descriptors has yet to be reported. In this project you explore the use of archetypal analysis to find the pure types of gold nanoparticles using a set of different descriptors, with more of less comprehensive information.  The data sets will be provided.


Determine how the choice of descriptor affects the identification of metallic nanoparticle archetypes.


Python programming and experience in data science and machine learning is essential (such as COMP3720, COMP4660, COMP4670, COMP6670, COMP8420).  Familiarity with platforms such as scikit-learn, Pytorch, Tensorflow and Keras is desirable.


This is a 12cp project.


machine learning, materials informatics, archetypes

Updated:  1 June 2019/Responsible Officer:  Dean, CECS/Page Contact:  CECS Marketing