Dark Mode
Superconductor Feature Engineering Data
Data Science and Analytics
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
This data product enables researchers and modellers to build predictive models for determining a superconductor's critical temperature based purely on its elemental and structural features. It includes raw data derived from public scientific sources, offering a valuable resource for studying the properties that govern superconductivity. The primary goal is to facilitate accurate predictive analysis for material discovery and characterisation.
Columns
The dataset includes 81 distinct feature attributes, along with the critical temperature (Tc) as the label column, totalling 82 columns in the main file (
train.csv). The features are derived characteristics of the superconducting material's constituent elements.Key attributes include:
number_of_elements: The count of chemical elements present.mean_atomic_mass: The average atomic mass of the constituent elements.wtd_mean_atomic_mass: The weighted average atomic mass.gmean_atomic_mass: The geometric mean of atomic mass.wtd_gmean_atomic_mass: The weighted geometric mean of atomic mass.entropy_atomic_mass: The entropy related to atomic mass.range_atomic_mass: The range of atomic mass values.std_atomic_mass: The standard deviation of atomic mass.
A separate file (
unique_m.csv) contains the chemical formula broken down for each material.Distribution
The dataset is structured into two main files, typically delivered in CSV format. The primary file,
train.csv, contains 21,263 records/superconductors and occupies approximately 23.86 MB. The data is notably clean, showing 100% validity and 0% missing values across the sampled features.Usage
This data is ideally suited for developing advanced statistical models and machine learning algorithms, such as regression models, intended to predict the critical temperature of new or existing materials. It serves as foundational training data for academic research in condensed matter physics, materials informatics, and data science competitions focused on physical property prediction.
Coverage
The scope of this dataset covers 21,263 distinct superconducting compounds. The data is physics-centric, focusing on the properties of the material composition and structure. Specific geographical, time range, or demographic scope limitations are not applicable to this physical science data.
License
CC0: Public Domain
Who Can Use It
- Materials Scientists: For validating theories on superconducting properties and material composition.
- Data Scientists/Machine Learning Engineers: For training and evaluating predictive models (e.g., critical temperature regression).
- Researchers in Physics: Studying the relationship between elemental attributes and superconductivity onset.
Dataset Name Suggestions
- Superconductivity Critical Temperature Prediction Dataset
- Superconductor Feature Engineering Data
- Materials Science Tc Regression Data
- Data-Driven Superconductor Model Features
Attributes
Original Data Source: Superconductor Feature Engineering Data
QUALITYLoading...