data-preprocessing

Automate data cleaning, missing-value handling, outlier detection, and correlation analysis.

1|Updated Apr 2, 2026
One-click install
npx skills add https://github.com/Best6668/AMIS --skill data-preprocessing
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: data-preprocessing
Source: https://github.com/Best6668/AMIS/tree/main/skills/data-preprocessing
Command: npx skills add https://github.com/Best6668/AMIS --skill data-preprocessing

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

数据预处理在数据科学和建模工作流中常常耗时且易出错。本 Skill 提供端到端的数据清洗、缺失值处理、异常值检测、分布与相关性分析等能力,帮助用户快速从原始数据获得可直接用于建模的高质量数据和可解释的质量评估结果。

Core Features & Use Cases

  • 数据加载与编码检测:自动识别数据文件和编码,稳定进入分析管线。
  • 缺失值分析与处理策略:统计缺失率,区分 MCAR/MAR/MNAR,并给出数值型/分类型的填充建议和处理工作流。
  • 异常值检测与治理:同时使用 Z-score 与 IQR 标准识别强弱异常,给出分类与保留/修正的决策指引。
  • 分布与相关性分析:绘制分布、计算偏度/峰度、得到 Pearson/Spearman 相关性矩阵并标注高相关对。
  • 数据清洗输出:生成数据清洗日志、 cleaned 数据集与质量报告,便于后续建模使用。

Quick Start

将数据集放置在 data/ 目录后,执行数据预处理技能以生成清洗后的数据和质量分析报告。

Frequently Asked Questions about data-preprocessing

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate data cleaning and missing value handling for a structured dataset?

You can automate data cleaning by using this skill to detect missing values, classify them as MCAR, MAR, or MNAR, and apply type-specific imputation. It outputs a cleaned dataset and a quality report ready for modeling pipelines.

What is the best way to detect outliers using Z-score and IQR in pandas?

Detect outliers by applying both Z-score and IQR standards simultaneously to identify strong and weak anomalies. This skill provides classification and decision guidance on whether to retain or correct the detected outliers.

How does correlation analysis work for preparing a dataset for modeling?

Correlation analysis for modeling works by calculating Pearson and Spearman matrices and flagging highly correlated variable pairs. This skill computes skewness and kurtosis to evaluate distributions before feeding data into a model.

Do I need to manually specify file encoding when loading raw data for preprocessing?

No, you do not need to manually specify file encoding. The skill automatically identifies data files and their encodings to ensure stable ingestion into the analytics pipeline without manual intervention.

What libraries are required to generate data quality reports and visualizations?

Generating data quality reports and visualizations requires pandas, numpy, and visualization libraries. These dependencies power distribution plotting, correlation matrices, and standardized artifact generation for modeling pipelines.