Classification Algorithms

Train, evaluate, and compare 9 supervised classification models on your data

Algorithms Available

Train/Test

Stratified Splitting

15+

Metrics & Charts

CSV / XLSX

Upload Your Data

Getting Started

Quick Guide

Upload Your Data

Upload a CSV or Excel file with features and a target column. Non-numeric columns are auto-encoded.

Pick an Algorithm

Choose from 9 classifiers and configure parameters like train/test split, regularization, and more.

Analyze Results

Get confusion matrix, ROC curves, feature importance, precision/recall and many more metrics.

Classification Algorithms

Choose an algorithm based on your data characteristics and interpretability needs

Linear

2 algorithms

Available

Logistic Regression

The Probabilistic Baseline

A fundamental classification algorithm that models the probability of class membership using a logistic function. Excellent interpretability through coefficients and calibrated probabilities.

Key Features

ProbabilisticInterpretableCoefficientsCalibration CurvesRegularization

Best for: Binary/multiclass with interpretable results

Try it now

Available

SVM

Maximum Margin Classifier

Support Vector Machine finds the optimal hyperplane that maximizes the margin between classes. With kernel tricks, it can handle non-linear decision boundaries.

Key Features

Max-marginKernel trickRobustSupport vectorsHigh-dimensional

Best for: High-dimensional data, text classification

Try it now

Tree-Based

4 algorithms

Available

Decision Tree

Intuitive Rule-Based

Builds a tree of decision rules that split data by features. Highly interpretable — you can trace any prediction back to a series of simple if-then conditions.

Key Features

InterpretableFeature importanceNo scaling neededHandles mixed types

Best for: When interpretability is critical

Combines many decision trees trained on random subsets of data and features. Reduces overfitting through bagging and provides robust predictions.

Key Features

EnsembleBaggingLow overfittingFeature importanceRobust

Best for: General-purpose, robust classification

Builds trees sequentially, with each new tree correcting errors from previous ones. Achieves high accuracy through gradient-based optimization.

Key Features

BoostingHigh accuracyStaged learningLearning rate control

Best for: Structured/tabular data competitions

Try it now

Available

XGBoost

Optimized Gradient Boosting

An optimized implementation of gradient boosting with regularization, parallel processing, and built-in handling of missing values. Industry standard for tabular data.

Key Features

RegularizedFastMissing value handlingFeature importanceIndustry standard

Best for: Production ML, Kaggle competitions

Try it now

Probabilistic

1 algorithm

Available

Naive Bayes

Bayesian Classifier

Applies Bayes' theorem with a "naive" independence assumption between features. Surprisingly effective and extremely fast, especially for text classification.

Key Features

Very fastProbabilisticText-friendlyLow data requirement

Best for: Text classification, spam filtering

Try it now

Instance-Based & Neural

2 algorithms

Available

KNN

K-Nearest Neighbors

Classifies new data points based on the majority vote of their k nearest neighbors. Non-parametric and intuitive — no training phase, just distance computation.

Key Features

Non-parametricNo trainingDistance-basedAdaptive boundsWeighted voting

Best for: Small to medium datasets, recommendation systems

Try it now

Available

MLP

Neural Network Classifier

Multi-Layer Perceptron — a feedforward neural network with configurable hidden layers. Learns complex non-linear decision boundaries through backpropagation.

Key Features

Deep learningNon-linearMulti-layerLoss curvesWeight visualization

Best for: Complex patterns, image-like features

Try it now

Need Help Choosing?

Start with Logistic Regression for a quick interpretable baseline, then try Random Forest or XGBoost for higher accuracy.

Start with Logistic Regression

Loading…