A.K. Cybulska

Translating Granularity of Event Slots into Features for Event Coreference Resolution.

Using clues from event semantics to solve coreference, we present an “event template” approach to cross-document event coreference resolution on news articles. The approach uses a pairwise model, in which event information is compared along five semantically motivated slots of an event template. The templates, filled in on the sentence level for every event mention from the data set, are used for supervised classification. In this study, we determine granularity of events and we use the grain size as a clue for solving event coreference. We experiment with a newly-created granularity ontology employing granularity levels of locations, times and human participants as well as event durations as features in event coreference resolution. The granularity ontology is available for research. Results show that determining granularity along semantic event slots, even on the sentence level exclusively, improves precision and solves event coreference with scores comparable to those achieved in related work.

Identity and Granularity of Events in Text

In this paper we describe a method to detect event descriptions in different news articles and to model the semantics of events and their components using RDF representations. We compare these descriptions to solve a cross-document event coreference task. Our component approach to event semantics defines identity and granularity of events at different levels. It performs close to state-of-the-art approaches on the cross-document event coreference task, while outperforming other works when assuming similar quality of event detection. We demonstrate how granularity and identity are interconnected and we discuss how semantic anomaly could be used to define differences between coreference, subevent and topical relations.

How concrete do we get telling stories

Will reading different stories about the same event in the world result in a similar image of the world? Will reading the same story by different people result in a similar proxy for experiencing the story? The answer to both questions is no because language is abstract by definition and relies on our episodic experience to turn a story into a more concrete mental movie. Since our episodic knowledge differs, also the mental movie will be different. Language leaves out details, and this becomes specifically clear when building machines that read texts to represent events and to establish event relations across mentions, such as co-reference, causality, subevents, scripts, timelines, and storylines. There is a lot of information and knowledge on the event that is not in the text but is needed to reconstruct these relations and understand the story. Machines lack this knowledge and experience and likewise make explicit what it takes to understand stories from text. In this paper, we report on experiments to automatically model event descriptions and instances across different news articles. We will show that event information is scattered over the text but also varies a lot in the degree it abstracts from details, which makes establishing event identity and relations extremely difficult. The variation in granularity of event descriptions seems to vary with pragmatic communicative strategies and defines the problem at different levels of complexity.