Superbio AutoML: Classification Workflow provides an intuitive, no-code approach to building machine learning models for classifying data into distinct categories. Whether working with binary or multiclass problems, this workflow leverages robust algorithms like XGBoost, ensemble methods, and bagging to identify the best-performing model tailored to your dataset.

Users can effortlessly upload CSV data, define their target variable, and select evaluation metrics such as the Matthews Correlation Coefficient (MCC) for binary classification. The platform ensures transparency by allowing users to set the test data split percentage, making it easy to gauge the model's real-world performance.

To achieve optimal results, the workflow emphasizes the importance of clean, leakage-free datasets. With a focus on simplifying the complexities of machine learning, this AutoML workflow empowers users to develop powerful classification models efficiently, regardless of their technical expertise.

Limitations: Data should not have leakage and be as clean as possible.

Technology: Classical ML models including XgBoost, Ensamble, and Bagging.

Use cases: Create a classification model (binary or multiclass) from tabular data.