Lightweight Boosting Models for User Response Prediction Using Adversarial Validation

TL;DR

This paper presents a lightweight, effective approach for user response prediction in app install prediction tasks, utilizing adversarial validation and gradient boosting, achieving competitive results in a challenge setting.

Contribution

The paper introduces a lightweight solution combining adversarial validation, feature engineering, and LightGBM for user response prediction, emphasizing simplicity and efficiency.

Findings

Single LightGBM model performs well without ensembling

Effective feature selection via adversarial validation

Achieved 9th place in RecSys Challenge 2023

Abstract

The ACM RecSys Challenge 2023, organized by ShareChat, aims to predict the probability of the app being installed. This paper describes the lightweight solution to this challenge. We formulate the task as a user response prediction task. For rapid prototyping for the task, we propose a lightweight solution including the following steps: 1) using adversarial validation, we effectively eliminate uninformative features from a dataset; 2) to address noisy continuous features and categorical features with a large number of unique values, we employ feature engineering techniques.; 3) we leverage Gradient Boosted Decision Trees (GBDT) for their exceptional performance and scalability. The experiments show that a single LightGBM model, without additional ensembling, performs quite well. Our team achieved ninth place in the challenge with the final leaderboard score of 6.059065. Code for our approach can be found here: https://github.com/choco9966/recsys-challenge-2023.