Loess Compressibility Model Predictions for Xinjiang Study Areas

5.5 KB of data compares multilayer perceptron neural network model predictions for loess compressibility with measured data. The dataset includes error metrics and covers typical study areas in Huocheng and Xinyuan counties, Xinjiang. It was authored by Zhiqi Liu and published under a CC BY 4.0 license.

Use Cases

• Compare loess compressibility parameter predictions from MLP and Random Forest models using provided error metrics.
• Analyze correlations between core input indicators and loess compressibility parameters like the compressibility modulus.
• Validate SVM classification or CART model performance for engineering classification of mechanical parameters in the study areas.

Strengths

• Dataset is openly licensed under CC BY 4.0, enabling broad reuse.
• Focuses on specific study areas in Huocheng and Xinyuan counties, Xinjiang, providing geographic context.
• Includes comparisons of multiple model types (MLP, Random Forest, SVM, CART) for loess compressibility.

Limitations

• Very small dataset at only 5.5 KB, indicating limited scope and likely few data points.
• No column names or sample data are provided, making structure and content assessment difficult.
• The description is highly technical and repetitive, lacking clear details on data volume and specific variables.

Provenance

Source

Zhiqi Liu via figshare

Collection Method

null

Time Range

null

Freshness

null

Geography

Huocheng and Xinyuan counties, Xinjiang, China