zoomzoom1011 / Auto-Encoder-Decoder-and-Embedding-Framework-in-Click-Prediction

Auto-Encoder-Decoder-and-Embedding-Framework-in-Click-Prediction

Nowadays, predicting click-through rate is important for gaining business insight and increase benefit. Here, I performed analysis on Click-Through data in Python to predict Click-Through rate.

This project includes:

1. Built and trained a wide and deep learning model with hidden layers and embedding layers
2. Built an auto-encoder-decoder in PySpark
3. Used models(DeepFM, wide and deep, xDeepFM) from DeepCTR and select the best model based on their performance on test datasets