Loess Geotechnical Parameter Statistical Analysis Results

Zhiqi Liu's dataset contains probabilistic statistical analysis results for loess geotechnical parameters, shared under a CC BY 4.0 license. The analysis, last updated in March 2026, includes results from models like Random Forest, MLP, CART, and SVM. The dataset is small, stored in a 9.5 KB XLS file.

Use Cases

• Compare error metrics priority across Random Forest, MLP, CART, and SVM classification models for loess parameter prediction.
• Analyze correlations, such as significantly positively correlated or negatively correlated relationships, between loess compressibility parameters and other mechanical parameters.
• Apply the engineering classification scheme to loess data from typical study areas like Huocheng and Xinyuan counties in Xinjiang.
• Use the results of performed normality tests to inform probabilistic modeling of the compressibility modulus.

Strengths

• Dataset is openly licensed under CC BY 4.0, facilitating reuse and sharing.
• Analysis incorporates multiple machine learning models (Random Forest, MLP, CART, SVM) for comparative results.
• Focuses on specific geotechnical parameters for loess, including compressibility and mechanical parameters.
• Geographic scope is defined for typical study areas in Huocheng and Xinyuan counties, Xinjiang.

Limitations

• Extremely small data scale at 9.5 KB, indicating it contains summary results, not the underlying raw data for model training.
• No column definitions or sample data are provided, making the internal structure and specific variables unclear.
• The description is very brief and lacks details on sample size, data collection methodology, or temporal coverage.

Provenance

Source

Zhiqi Liu via figshare

Collection Method

Probabilistic statistical analysis of loess geotechnical parameters, likely from engineering studies.

Freshness

Last updated March 2026.

Geography

Typical study areas include Huocheng and Xinyuan counties in Xinjiang, China.