svm

Vegetation/non-vegetation land cover classification of Medellín's Green Corridors using Sentinel-2, Random Forest and SVM. Includes multi-temporal change detection and Landsat-8 LST analysis to assess urban reforestation impact.

A web application to display and predict market trends for certain stocks.

Heart disease risk prediction — Django + SVM, deployable on Render

human detection system

Financial sentiment analysis comparing LR, SVM and DistilBERT with backtesting against real stock prices

Persistent agent identity and on-chain memory SDK for the X1 blockchain

Cookie Crumbs - fully on-chain tip jar cApp on Cookie Chain (SVM). Community tip splits, goals, embed widget. Republication after altaranexus-ship-it suspension.

Canonical public chain metadata for Rome Protocol — chains, contracts, tokens, and program ids; read it instead of hardcoding.

Generate custom vanity addresses on Solana and EVM chains with AWS KMS-secured minting

This APP allows you to classify a Lansat-8 image with Random Forest, Cart or Svm with the best HyperParameters with and without texture

Collaborative pixel board where every pixel is a real on-chain transaction (memo CCv1 + tiny transfer) — the whole artwork rebuilds from raw chain history. Cookie Chain + Solana Devnet, one toggle. Live: https://loveoftheai.github.io/cookie-canvas/

Breast cancer detection is a web application to detect breast cancer in women using parametric data. Backend is built using machine learning and Django. Bootstrap is used for frontend.

Breast Cancer Detection. This is a project based on Machine Learning. It is Full stack, created using Django [Python]. Used for detecting Breast Cancer from parametric data imported from sklearn.

ML-based beta thalassemia carrier screening from CBC parameters for Sri Lankan primary care. SVM + SHAP explainability, MOH referral letters, couple screening, voice input, and analytics dashboard.

ML algorithms visualized in vanilla canvas — zero dependencies

Open-source forensic transaction reproduction infrastructure for Solana

I developed a computer-aided detection system (CAD) using machine learning techniques, including Grayscale, Median filter, High-pass filter, Otsu's Thresholding Technique, and PCA Feature Extraction. The hybrid classifier (SVM-KNN) improved accuracy, and the system identifies brain tumors with 98% accuracy.