Paper: Recognizing Identical Events with Graph Kernels

ACL ID P13-2139
Title Recognizing Identical Events with Graph Kernels
Venue Annual Meeting of the Association of Computational Linguistics
Session Short Paper
Year 2013
Authors

Identifying news stories that discuss the same real-world events is important for news tracking and retrieval. Most exist- ing approaches rely on the traditional vec- tor space model. We propose an approach for recognizing identical real-world events based on a structured, event-oriented doc- ument representation. We structure docu- ments as graphs of event mentions and use graph kernels to measure the similarity be- tween document pairs. Our experiments indicate that the proposed graph-based ap- proach can outperform the traditional vec- tor space model, and is especially suitable for distinguishing between topically simi- lar, yet non-identical events.