Superconductor Critical Temperature Predictions for Regression Tasks

A tabular dataset curated for evaluating predictive machine learning models on independent and identically distributed data. The intended task is regression, specifically predicting the critical temperature of superconductors. The data originates from a 2018 study published in Computational Materials Science.

Use Cases

• Benchmarking regression model performance on independent and identically distributed tabular data
• Predicting the critical temperature of superconductors based on material features
• Studying model behavior on datasets with a high degree of duplicate feature vectors

Strengths

• Dataset has a clear, peer-reviewed source with a citation to a 2018 scientific paper
• Licensed under CC BY 4.0 for open use and redistribution
• Curated specifically for structured evaluation in a machine learning study

Limitations

• Column-level documentation is absent; field semantics must be inferred after download
• Row count is unknown, which may limit suitability assessment
• The data contains a high rate of duplicate feature vectors, reported as 29%

Provenance

Source

https://doi.org/10.24432/C53P47 and Hamidieh, Kam. 'A data-driven statistical model for predicting the critical temperature of a superconductor.' Computational Materials Science 154 (2018): 346-354.

Collection Method

Curated from the original source by the TabArena team for a tabular ML study.

Time Range

2018

Freshness

Dataset Year is 2018; last update date is unknown.

Geography

null