What algorithms are included in this classification skill?

The skill covers a wide range of algorithms including Logistic Regression, Decision Trees, Random Forest, Gradient Boosting, SVM, and Naive Bayes.

What evaluation metrics are provided?

It generates Accuracy, Precision, Recall, F1-Score, AUC-ROC curves, and Confusion Matrices to ensure a holistic view of model performance.

How does this skill help with imbalanced data?

It provides implementation patterns for oversampling, undersampling, SMOTE (Synthetic Minority Oversampling Technique), and the application of class weights.

Can I use this for multi-category classification?

Yes, the skill supports Binary Classification (two classes), Multiclass (more than two), and Multi-label (multiple classes per observation) scenarios.

Classification Modeling

Name: Classification Modeling
Author: aj-geddes

byaj-geddes

•

Data Science & ML

Build and evaluate robust binary and multiclass classification models using industry-standard machine learning algorithms and metrics.

The Classification Modeling skill provides a comprehensive framework for developing predictive models that assign observations to discrete categories. It streamlines the implementation of various algorithms—ranging from Logistic Regression to advanced ensemble methods like Random Forest and Gradient Boosting—while providing built-in tools for rigorous evaluation. Whether you are tackling binary outcomes like churn prediction or complex multiclass tasks, this skill offers the patterns needed for feature scaling, handling class imbalances, and generating detailed performance visualizations like ROC curves and confusion matrices.

Key Features

01Advanced techniques for handling class imbalance using SMOTE and weighted penalties.

02Implementation of popular algorithms including Logistic Regression, Random Forest, and Gradient Boosting.

03Standardized workflows for cross-validation and model comparison.

04Comprehensive evaluation suite featuring AUC-ROC, F1-Score, and Precision-Recall metrics.

05Automated visualization tools for confusion matrices, feature importance, and probability calibration.

0618 GitHub stars

Use Cases

01Automating fraud detection and anomaly identification in production systems.

02Predicting customer churn and loan default risk in financial services.

03Classifying sentiment or product categories from user-generated content.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add aj-geddes/useful-ai-prompts classification-modeling

For use in Claude.ai and ChatGPT

Download Skill