gvm-analysis

Privacy-preserving exploratory data analysis analyzes structured CSV files and generates aggregated findings reports without exposing raw data. It supports formats like TSV, xlsx, parquet, and JSON, ensuring privacy while providing insights into dataset shape, outliers, and trends.

Reproducible data analysis reports include provenance information and a methodology appendix for tracking. The skill generates grounded findings reports with insights, ensuring that the analytical process remains transparent and reproducible across different datasets.

The exploratory data analysis tool supports parquet, xlsx, CSV, TSV, and JSON formats. It leverages Python libraries like pandas, numpy, and scipy to process these structured data files and extract aggregated findings without exposing underlying raw records.

To understand dataset outliers and trends without writing code, run an automated exploratory data analysis skill. It processes your structured data files and outputs a grounded findings report containing aggregated insights, statistical summaries, and detected trends.

Privacy-preserving data analysis works by using statistical libraries like statsmodels and scikit-learn to compute aggregated findings. It ensures no raw data is exposed in the analysis process, outputting only grounded insights and statistical summaries in the final report.

Limitations when generating reports from large datasets depend on the memory constraints of the pandas and numpy Python libraries. The skill requires these dependencies for statistical analysis and data processing, ensuring aggregated findings are extracted without exposing raw data.