Detecting Relevant Feature Interactions for Recommender Systems via Graph Neural Networks

08/02/2020
by   Yixin Su, et al.
16

Feature interactions are essential for achieving high accuracy in recommender systems (RS), so they have been taken into consideration in many existing RS, where all feature interactions are modeled. Nevertheless, not all feature interactions have positive effects for RS: modeling the irrelevant feature interactions may introduce noises and degrade the accuracy. To overcome this problem, in this work, we propose a graph neural network-based model, L0-SIGN, to detect the relevance of feature interactions and utilize only the relevant ones for RS, with features as nodes and feature interactions as edges. Generally, our model consists of two components: an L0 regularization based edge prediction model to explicitly detect relevant feature interactions; and a graph classification model, SIGN, to effectively model and aggregate the detected ones for recommendations. These two components positively influence each other to ensure that the most relevant feature interactions will be detected and modeled. In addition, we further prove that the effectiveness of our model is theoretically sound. We first show that our model is a variational approximation of information bottleneck principle, i.e., the detected feature interactions are guaranteed to be most relevant. We then show that our model follows the definition of statistical interactions, proving that the modeling of detected feature interactions in L0-SIGN is effective. Experimental results show that (i) L0-SIGN outperforms existing baselines in terms of accuracy, and (ii) the detected feature interactions are beneficial for performance gain and interpretability.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
10/12/2019

Fi-GNN: Modeling Feature Interactions via Graph Neural Networks for CTR Prediction

Click-through rate (CTR) prediction is an essential task in web applicat...
research
05/25/2022

Factorized Structured Regression for Large-Scale Varying Coefficient Models

Recommender Systems (RS) pervade many aspects of our everyday digital li...
research
08/19/2021

SiReN: Sign-Aware Recommendation Using Graph Neural Networks

In recent years, many recommender systems using network embedding (NE) s...
research
03/05/2021

Graph Convolutional Embeddings for Recommender Systems

Modern recommender systems (RS) work by processing a number of signals t...
research
07/21/2021

Structure-aware Interactive Graph Neural Networks for the Prediction of Protein-Ligand Binding Affinity

Drug discovery often relies on the successful prediction of protein-liga...
research
05/20/2022

Tackling Provably Hard Representative Selection via Graph Neural Networks

Representative selection (RS) is the problem of finding a small subset o...
research
12/20/2017

On the Relation of External and Internal Feature Interactions: A Case Study

Detecting feature interactions is imperative for accurately predicting p...

Please sign up or login with your details

Forgot password? Click here to reset