Auto-Detection of Safety Issues in Baby Products

TL;DR

This paper presents an automated system that analyzes online reviews and safety complaints to detect potential safety issues in baby products, achieving high precision and outperforming existing benchmarks.

Contribution

The study introduces a novel approach combining text processing, feature selection, and machine learning classifiers to identify safety concerns from consumer reviews and complaint data.

Findings

Logistic regression achieved 66% precision in top safety issue detection.

The proposed system outperforms existing benchmarks and expert assessments.

Feature relevance was effectively determined using Random Forests.

Abstract

Every year, thousands of people receive consumer product related injuries. Research indicates that online customer reviews can be processed to autonomously identify product safety issues. Early identification of safety issues can lead to earlier recalls, and thus fewer injuries and deaths. A dataset of product reviews from Amazon.com was compiled, along with \emph{SaferProducts.gov} complaints and recall descriptions from the Consumer Product Safety Commission (CPSC) and European Commission Rapid Alert system. A system was built to clean the collected text and to extract relevant features. Dimensionality reduction was performed by computing feature relevance through a Random Forest and discarding features with low information gain. Various classifiers were analyzed, including Logistic Regression, SVMs, Na{\"i}ve-Bayes, Random Forests, and an Ensemble classifier. Experimentation with various features and classifier combinations resulted in a logistic regression model with 66\% precision in the top 50 reviews surfaced. This classifier outperforms all benchmarks set by related literature and consumer product safety professionals.