Pinned Loading

1. bike-sharing-demand-prediction bike-sharing-demand-prediction Public

   Predicting hourly bike rental demand using LightGBM, XGBoost, and Random Forest on the Capital Bikeshare dataset. Includes full EDA, model comparison, and evaluation with RMSE, MAE, and R².

   Jupyter Notebook

2. YOGA-Chatbot YOGA-Chatbot Public

   YOGA (YOgyakarta Guide Assistant) — A Telegram chatbot helping tourists discover destinations across Yogyakarta, powered by a hybrid SVM NLP pipeline with 97.33% accuracy.

   Jupyter Notebook

3. by.u-sentiment-analysis by.u-sentiment-analysis Public

   Comprehensive sentiment analysis of 227,588+ by.U mobile app reviews using NLP and machine learning. Includes data preprocessing, word cloud visualizations, and model comparison.

   Jupyter Notebook

4. car-price-prediction car-price-prediction Public

   End-to-end machine learning pipeline for predicting used car prices using the CarDekho dataset — EDA, feature engineering, and multi-model benchmarking.

   Jupyter Notebook

5. chest-xray-pneumonia-detection chest-xray-pneumonia-detection Public

   Binary classification of chest X-ray images (Normal vs Pneumonia) using a custom CNN with TensorFlow/Keras. Achieves 0.996 AUC with optimal decision threshold tuning on 5,856 images.

   Jupyter Notebook

6. pinjol-app-sentiment-analysis pinjol-app-sentiment-analysis Public

   Sentiment analysis of Indonesian P2P lending (Pinjaman Online) app reviews from Google Play Store using TF-IDF and Word2Vec with 10 classification models. Best result: BiLSTM + Word2Vec at 98.72% a…

   Jupyter Notebook