Increasing mining activity requires efficient land restoration. While traditional methods are costly and
disruptive, phytoremediation is a sustainable, plant-based method for remediating contaminated soils.
However, selecting the most efficient plant species for specific contaminants, soil types, and climate
zones is often challenging due to the large number of environmental variables.

This project seeks to streamline and optimize phytoremediation by developing a machine learning
driven recommendation system based on an automated literature review. With this, we aim to predict
the most suitable plant species based on the target element(s) to be removed from the soil, as well as
soil characteristics and climate zone, in order to enhance phytoremediation strategies