# Feature Generation by Convolutional Neural Network for Click-Through Rate Prediction

## TL;DR

This paper introduces FGCNN, a novel CNN-based feature generation method that automatically augments feature space to improve click-through rate prediction, outperforming existing models on large datasets.

## Contribution

The paper proposes a new FGCNN model combining feature generation via CNN with a deep classifier, reducing reliance on manual feature engineering for CTR prediction.

## Key findings

- FGCNN significantly outperforms nine state-of-the-art models on large datasets.
- The model is highly compatible with various deep classifiers.
- Automatic feature generation improves DNN learning efficiency.

## Abstract

Click-Through Rate prediction is an important task in recommender systems, which aims to estimate the probability of a user to click on a given item. Recently, many deep models have been proposed to learn low-order and high-order feature interactions from original features. However, since useful interactions are always sparse, it is difficult for DNN to learn them effectively under a large number of parameters. In real scenarios, artificial features are able to improve the performance of deep models (such as Wide & Deep Learning), but feature engineering is expensive and requires domain knowledge, making it impractical in different scenarios. Therefore, it is necessary to augment feature space automatically. In this paper, We propose a novel Feature Generation by Convolutional Neural Network (FGCNN) model with two components: Feature Generation and Deep Classifier. Feature Generation leverages the strength of CNN to generate local patterns and recombine them to generate new features. Deep Classifier adopts the structure of IPNN to learn interactions from the augmented feature space. Experimental results on three large-scale datasets show that FGCNN significantly outperforms nine state-of-the-art models. Moreover, when applying some state-of-the-art models as Deep Classifier, better performance is always achieved, showing the great compatibility of our FGCNN model. This work explores a novel direction for CTR predictions: it is quite useful to reduce the learning difficulties of DNN by automatically identifying important features.