A股量化选股系统 基于LightGBM
XGBoost, LightGBM, LSTM, Linear Regression, Exploratory Data Analysis
Code repo of solution of 11th place in Recsys Challenge 2022
predict the winning horse with supervised machine learning models (lucky to have 100% accuracy on small test data)
Advancing Healthcare with 91% Accurate Prediction of Obesity Risk Levels Using XGBoost ,LightGBMand CatBoostClassifier Model
Location information about commuter activities is vital for planning for travel disruptions and infrastructural development. The Mobility Sensing Project aims to find innovative and novel ways to identify travel patterns from GPS data and other multi-sensory data collected in smartphones. This will be transformative to provide personalised trave…
[Information System] SMUTF: Schema Matching Using Generative Tags and Hybrid Features
AI powered retail intelligence platform for demand forecasting, inventory optimization, explainable AI, and automated business insights, built with React, FastAPI, Python, MySQL, Redis, Celery, and Docker.
Southern Data Science Conference Attempt 2020
AI-driven retention platform "TalentGuard". Predicting high-performer attrition 2 months in advance with 0.84 AUC. / AI離職予測プラットフォーム「TalentGuard」。AUC 0.84の高精度予測でエース人材の離職を2ヶ月前に検知。
End-to-end quantitative trading pipeline with feature engineering, ensemble ML models, and modular portfolio backtesting.
End-to-end Data Science project — Sales Analytics + LightGBM forecasting model + Interactive Plotly Dash dashboard for Rossmann Store Sales.
Personalized Hybrid Recommender (User-to-Item) for games, utilizing SVD, TF-IDF, and ensemble methods on 230K+ reviews. Features integrated MongoDB data pipeline, sentiment analysis, and a live Streamlit dashboard.
Autism Prediction in Adults
M5 Forecasting データセットを用いた 実務向け需要予測パイプラインです。
Developer Salary prediction is an AI Competition in Quevico, It bases on the survey of Stackoverflow. More detail is available in the link (https://quevico.com/competitions/105)
Business-oriented fraud detection using time-based validation and operational metrics.
Flight Price Prediction Model Deployment IN Heroku
Tabular Playground Series - May 2022
End-to-end credit risk ML system — GradientBoosting vs LightGBM vs Statsmodels comparison, calibrated probabilities, Docker containerized, AWS S3/EC2, CI/CD via GitHub Actions, numpy-only serverless inference on FastAPI/Vercel, React dashboard.
To associate your repository with the lightgbm-models topic, visit your repo's landing page and select "manage topics."