Practice Projects of Machine Learning.....

Credit Card Fraud Detection using Machine Learning, Random Forest, SMOTE, and an interactive Streamlit dashboard to identify potentially fraudulent transactions.

GSoC organization analytics tool that scrapes historical participation data, exports structured Excel datasets, and predicts next-year organization presence using machine learning.

Reference-assisted evaluation of a granular-computing matting pipeline for mammogram masses (CBIS-DDSM, MIAS), with seed-only controls and paired bootstrap statistics.

Learn how apps guess what you will like: similarity, prediction, matrix factorisation and fair testing, in 3 sessions.

An empirical comparison of deterministic vs. stochastic exploration in stationary Multi-Armed Bandits.

Real-time fraud detection — sliding-window velocity features, impossible travel, spend z-scores, explainable rule DSL with hot reload, online logistic model, review queue, HTTP scoring API, synthetic traffic with 4 fraud patterns. Live simulator. Prism Systems 09/12

A machine learning-based Rice Crop Yield Forecasting System that analyzes agricultural and environmental factors to predict rice yield and support data-driven farming decisions.

Practical, reproducible training and fine-tuning guides for embeddings, rerankers, LLMs and multimodal models. Learn with Jupyter notebooks, Google Colab and reusable Python modules.

An automated AI & ML news aggregator that clusters and ranks stories, generates AI briefings, and deploys a static news website twice daily using GitHub Actions.

Training of gradient-network magnetic models

Forecasting Swiss Trains' delay propagation

Customer income prediction for a digital bank (Bilkent IE 477/478 senior design project with Hayat Finans): R tidymodels notebook with OLS, LASSO, random forest, XGBoost and LightGBM, two Shiny apps, and a synthetic data generator so it runs without the confidential data.

Customer Shopping Trends & Behavior Analysis is to evaluate consumer purchasing behavior, segment customer using machine learning.

A neural network that learns to multiply, built from scratch in NumPy. Includes backprop, Adam, and an extrapolation experiment.

Stop water crises before they start. An early-warning and response system for city lakes and rivers, built on HL7 FHIR and the OneAquaHealth standard.

Interpretable 18-class computer-vision pipeline using U²-Net segmentation, handcrafted visual descriptors, Random Forest classification, and evaluation.

Data-leakage detector for tabular ML: static pipeline analysis, runtime provenance, point-in-time checks and statistical checks, with measured impact and controls. Architectured by Claude.

End-to-end Azure MLOps platform for fraud detection with Databricks, Azure ML, MLflow, GitHub Actions, CT/CD, drift monitoring, quality gates, blue/green deployment, rollback, and OIDC security.

Can outcome-trained routing choose how much evidence an AI reads? A local RAG experiment, and evaluation harnes.

A measurement-first telecom churn targeting platform combining machine learning, FastAPI scoring APIs, retention economics, and pilot sizing for data-driven customer retention.

A doctor for your ML training runs: reads TensorBoard/MLflow/W&B/stdout logs and diagnoses NaNs, divergence, overfitting, idle GPUs and memory leaks. CLI + skill for coding agents.

Predicts whether a reported adverse drug event is serious, with SHAP explanations. FDA FAERS + XGBoost + Streamlit.

Classifier that screens emergency micro-interventions for safety before physical testing. Simulated data.

How many more tokens (and dollars) the same text costs in your language. 222 languages, 7 LLM tokenizers, a paste-your-text playground, and an eval for your own tokenizer.

Flight delay analysis and leakage-free prediction on 31M US flights (Python, R, SQL)

Will this customer miss next month's payment? A credit default model, built end to end on 30,000 real cardholders.

Programming course for vibe coders: Python from zero to attention, in the browser

The classic algorithms, from scratch, for offline features: tf-idf + cosine, Naive Bayes, k-NN, k-means, PCA, ID3, linear regression, Markov chains and a grammar expander. Zero dependencies, single file, no model, no API key, no network. MIT.

Top Multi Cloud ML Orchestration (Opensource) Star if you like it!