The principle of the Parser can be as follows: 

·    Firstly, the Parser reads the chemical formula of an alloy (which is the first column in the mechanical property dataset). 

·    Secondly, it cleans redundant symbols in the chemical formula (such as spaces and brackets). 

·    Thereafter, it will normalise the proportion of different elements in the alloy composition. It will then provide the ratio for corresponding element and output a CSV file. The CSV file contains the 27 elements that appear in the dataset, along with the proportion of each element for each MPEA. 

·    The final output is utilised in machine learning processes.