Attention-Based Vandalism Detection in OpenStreetMap

TL;DR

This paper introduces Ovid, an attention-based neural network model that effectively detects vandalism in OpenStreetMap by leveraging novel features and a newly compiled dataset of vandalism incidents.

Contribution

The paper presents a novel neural architecture with multi-head attention for vandalism detection and provides the first open dataset of real-world vandalism in OSM.

Findings

Ovid outperforms existing methods in vandalism detection accuracy.

The new dataset enables better training and evaluation of vandalism detection models.

Attention mechanisms improve the summarization of vandalism indicators.

Abstract

OpenStreetMap (OSM), a collaborative, crowdsourced Web map, is a unique source of openly available worldwide map data, increasingly adopted in Web applications. Vandalism detection is a critical task to support trust and maintain OSM transparency. This task is remarkably challenging due to the large scale of the dataset, the sheer number of contributors, various vandalism forms, and the lack of annotated data. This paper presents Ovid - a novel attention-based method for vandalism detection in OSM. Ovid relies on a novel neural architecture that adopts a multi-head attention mechanism to summarize information indicating vandalism from OSM changesets effectively. To facilitate automated vandalism detection, we introduce a set of original features that capture changeset, user, and edit information. Furthermore, we extract a dataset of real-world vandalism incidents from the OSM edit history for the first time and provide this dataset as open data. Our evaluation conducted on real-world vandalism data demonstrates the effectiveness of Ovid.