Machine Learning for Heavy Metal Pollution Status Classification Using Spectroscopic and Electrochemical Data: A Systematic Review

This dataset supports the systematic review article "Machine Learning for Heavy Metal Pollution Status Classification Using Spectroscopic and Electrochemical Data: A Systematic Review", which synthesises evidence from 11 empirical studies on the application of machine learning and chemometric algorithms for classifying heavy metal pollution status in environmental matrices (water, soil, sediment, wastewater) using spectroscopic or electrochemical data. The dataset includes the PRISMA 2020 checklist, PRISMA flow diagram, conceptual framework figure (P‑E‑O model), criteria for inclusion and exclusion of studies (Table 1), databases and last search dates (Table 2), summary characteristics of the 11 included studies (authors, year, matrix, target metal(s), data type, ML algorithms) (Table 3), frequency of target heavy metals (Table 4), distribution of sample matrices (Table 5), spectroscopic techniques used (Table 6), electrochemical techniques used (Table 7), frequency of ML/chemometric algorithms (Table 8), common preprocessing and feature extraction methods (Table 9), summary of classification performance metrics (Table 10), integration of multiple data types across included studies (Table 11), risk of bias summary (Table 12), the data extraction form, and the JBI critical appraisal item‑by‑item assessment. All files are available under a Creative Commons Zero (CC0 1.0) license.