Clickbait detection: quick inference with maximum impact

TL;DR

This paper introduces a lightweight hybrid method for clickbait detection combining semantic embeddings and heuristic features, optimized for fast inference with competitive accuracy.

Contribution

It presents a novel hybrid approach that reduces embedding dimensionality and employs graph-based classifiers for efficient, accurate clickbait detection.

Findings

Graph-based models achieve high ROC-AUC with reduced inference time.

Simplified features slightly lower F1-scores but maintain strong detection performance.

Embedding reduction via PCA improves efficiency without significant accuracy loss.

Abstract

We propose a lightweight hybrid approach to clickbait detection that combines OpenAI semantic embeddings with six compact heuristic features capturing stylistic and informational cues. To improve efficiency, embeddings are reduced using PCA and evaluated with XGBoost, GraphSAGE, and GCN classifiers. While the simplified feature design yields slightly lower F1-scores, graph-based models achieve competitive performance with substantially reduced inference time. High ROC--AUC values further indicate strong discrimination capability, supporting reliable detection of clickbait headlines under varying decision thresholds.