ML system for prioritizing post-discharge follow-up using hospital readmission risk

Research-grade multimodal AI content detector using classical ML + neural fusion across text, code, and images — with XAI, uncertainty abstention, and a Rich TUI.

A small, local blockchain that records the provenance of machine-learning algs

A complete pipeline for classifying network traffic as Normal or as one of four attack categories (DoS, Probe, Privilege Escalation, Access) using a Random Forest classifier on the NSL-KDD dataset — with a proper train / validation / test split, per-class evaluation, and a saved, reusable model.

ANN-based estimated salary prediction system built with Python, TensorFlow, Scikit-learn, and Streamlit.

A small character-level GPT in PyTorch (LayerNorm, dropout, AdamW) with a CLI for training, sampling, and loss-curve evaluation.

LLM drift & quality monitoring: embedding drift, output-quality drift, and LLM-as-judge regression tracking

Reproducing and extending a Cryptography 2025 paper on Ethereum fraud detection (PSO + GA + XGBoost), with a leakage-safe benchmark and on-chain Solidity risk registry.

Network intrusion detection with PyTorch ANNs on UNSW-NB15 (10-class) with confidence-based alerting

Exploratory data analysis and a logistic regression model for loan approval and financial risk, using Python, pandas, seaborn and scikit-learn.

CASIA internship-derived traffic forecasting: Random Forest + PyTorch GRU, reproducible temporal evaluation, 12.4% lower GRU MAE, FastAPI inference.

Machine-learning prediction of InhA inhibitor potency from public ChEMBL data, with leakage checks.

Physics-Based Synthetic Coherent Diffraction Patterns for Phase Retrieval and Scientific Machine Learning

A grounded time-series forecasting framework that combines a language controller with specialist forecasting models. It turns natural-language requests into structured forecasting plans and evidence-grounded predictions, substantially outperforming general-purpose conversational language models used as direct forecasting baselines.

Sales prediction using python, pandas, and Linear Regression

Machine learning (MLOps) experiment tracking, hyperparameter tuning, model evaluation, model registry, and deployment using MLflow and Python.

面向 Obsidian 的本地论文阅读工作流：多源检索、证据筛选、AI 原文审校、中文笔记、双语 PDF、费用记录与断点恢复。

PySpark + Hive fraud pipeline on 10M card transactions: model vs rules, daily analyst alert queue with reasons

A complete Python machine learning pipeline leveraging Pandas, Scikit-Learn, and XGBoost for data cleaning, categorical encoding, exploratory data analysis, and predictive classification modeling.

Educational 20 Questions-style guessing game with shared learning

End-to-end ML swing trading system for XAUUSD & NAS100 — BOS event detection → XGBoost + calibration → EV filter → rule-based risk agent → MQL5 EA with staged stop-lock exits.

Semantic search over 253 Medium posts that runs AI in the browser. Transformers.js + all-MiniLM-L6-v2, a 1.4 MB vector index, no API key, no backend, works offline.

High-performance PostgreSQL C extension for local System 1 Decision Models (Julia-1, Lev, Kev) via llama.cpp. Direct SQL routing, calibrated probabilities, resume scoring, in-memory LRU cache, and million-row SRF batching.

Prepaid asynchronous LLM inference platform: FastAPI, PostgreSQL-backed queues, deadline-aware scheduling, modular GPU workers, vLLM integration, token metering, ledger billing, Stripe test payments, Next.js dashboard, Python SDK, and Docker. Includes a mock mode and recorded walkthrough; real GPU validation pending.

One-pass decision model built on Phi-4-mini-instruct that scores multiple-choice options without generating text.

Official ZEN and FOCUS code: Adapting Prior-Data Fitted Networks for Tabular Anomaly Detection (Technion)

Machine learning model (LightGBM) to forecast wholesale electricity prices (Day-Ahead Market) for the Greek national grid. Aggregates historical load data, DAM prices, and natural gas (TTF) prices in order to estimate future prices.

BMM4111 dönem ödevi: Sensör verileriyle sınıflandırma, regresyon ve kümeleme.