[Research] Transformer 기반 광고 기여도 모델 제안
-
Updated
Jan 17, 2025 - Jupyter Notebook
[Research] Transformer 기반 광고 기여도 모델 제안
AI-driven healthcare project for early prediction of conversion from Mild Cognitive Impairment (MCI) to Alzheimer's disease (AD). The system integrates OCR and NLP techniques to extract clinical information from handwritten medical records and leverages Machine Learning and Deep Learning models for neuroimaging-based analysis and prediction.
Large-scale e-commerce user behavior analysis and session-level purchase prediction with DuckDB, Parquet and LightGBM.
BA815 (BU Questrom, Fall 2025) team project on what drives digital marketing conversion: correlation, random forest and K-Means segmentation on an 8,000-customer campaign dataset.
A guided walkthrough of the development of an AI model that aims to predict whether a visitor session on an e-commerce website might result in a purchase. By analyzing historical session data, the project aims to enhance the user experience and increase conversion rates.
Predicting customer conversion in digital marketing campaigns with scikit-learn & XGBoost — feature selection, tuning, and a stacking ensemble (0.80 balanced accuracy).
Online shopping session conversion prediction and Streamlit experiment dashboard
Purchase-conversion prediction with theory-based CRM features: XGBoost holdout PR-AUC 0.3620 at 0.12% conversion
Machine learning classification pipeline (XGBoost, Random Forest) to predict e-commerce session conversion using behavioral clickstream data.
Session-level purchase intent with a business-value threshold, cost sensitivity sweep, decile lift, FastAPI serving and drift monitoring
Real-time e-commerce session conversion scoring — LSTM 0.9868 AUC · FastAPI 27ms · 121 tests · Live demo on HuggingFace
Comparative analysis of supervised learning for predicting conversion in an automotive case with special focus on imbalanced learning
An LSTM for ad conversion-rate prediction (pCVR) — last 50 user behavior events in, purchase probability out. Logistic regression baseline, strict time-based splits, FastAPI serving. 100M Taobao events.
Data science project analysing digital marketing data to identify customer conversion drivers and build predictive classification models.
To associate your repository with the conversion-prediction topic, visit your repo's landing page and select "manage topics."