Abstract
Subgroup discovery (SGD) is presented here as a data-mining approach to help find interpretable local patterns, correlations, and descriptors of a target property in materials-science data. Specifically, we will be concerned with data generated by density-functional theory calculations. At first, we demonstrate that SGD can identify physically meaningful models that classify the crystal structures of 82 octet binary (OB) semiconductors as either rocksalt or zincblende. SGD identifies an interpretable two-dimensional model derived from only the atomic radii of valence s and p orbitals that properly classifies the crystal structures for 79 of the 82 OB semiconductors. The SGD framework is subsequently applied to 24 400 configurations of neutral gas-phase gold clusters with 5–14 atoms to discern general patterns between geometrical and physicochemical properties. For example, SGD helps find that van der Waals interactions within gold clusters are linearly correlated with their radius of gyration and are weaker for planar clusters than for nonplanar clusters. Also, a descriptor that predicts a local linear correlation between the chemical hardness and the cluster isomer stability is found for the even-sized gold clusters.
Highlights
Rational design of advanced functional materials, e.g., active and selective catalysts [1], efficient thermoelectrics [2], and high-temperature superconductors [3], requires an understanding of the underlying fundamental physical mechanisms
The application of big-data analytics to obtain material insights and to predict novel materials can be enhanced by the availability of large materials repositories, e.g., AFLOWLIB, Computational Materials Repository, Electronic Structure Project, Materials Project, Novel Materials Discovery (NOMAD), Open Quantum Materials Database (OQMD), and Pauling file [26]
Our objective is to develop and exploit big-data analytics tools to discover materials insights and to predict advanced materials from large collections of materials data stored within the NOMAD Archive [27]
Summary
Rational design of advanced functional materials, e.g., active and selective catalysts [1], efficient thermoelectrics [2], and high-temperature superconductors [3], requires an understanding of the underlying fundamental physical mechanisms. In this paper we demonstrate a multipurpose data-mining algorithm called subgroup discovery (SGD) to identify and describe local patterns, correlations, and descriptors in materials-science data according to some desired target property (or properties) [29,30,31,32]. We demonstrate that SGD can identify physically meaningful models that classify the crystal structures of 82 octet binary semiconductors as either rocksalt (RS) or zincblende (ZB) from only information of its chemical composition. The aim of investigating gold clusters here is two-fold: (1) to search for general structure-property relationships holding across gold clusters of different sizes and vastly different configurations; and (2) to demonstrate the versatility of subgroup discovery on a large and heterogeneous dataset. It is established that SGD can help identify unexpected and general, size-independent, patterns within the dataset of gold cluster configurations
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
