It-TimeML is the adaptation of TimeML to Italian: it is a markup language for the annotation of temporal information in texts. The identification of temporal information is a critical component for any robust Natural Language Processing system, like Information Retrieval or language understanding systems, and recently it has been the topic of lots of research both in Computational Linguistics and Artificial Intelligence.
As for TimeML, It-TimeML focuses on Events (i.e. actions, states, and processes), Temporal Expressions (i.e. durations, calendar dates, times of day and sets of time), Signals (e.g. temporal prepositions and subordinators) and various kind of dependencies between Events and/or Temporal Expressions (i.e. temporal, aspectual and subordination relations).
Annotation specifications and guidelines have been developed for Italian and applied to the creation of the Ita-TimeBank, a language resource composed of two corpora manually annotated with temporal and event information. The first release of the Ita-TimeBank has been done on occasion of the EVENTI (EValuation of Events aNd Temporal Information) exercise in the context of Evalita 2014, the evaluation campaign for NLP and Speech Tools for Italian.
Goals
- Promote research in Temporal Processing for Italian
- Adapt TimeML to Italian
- Create an annotated language resource to train and test automatic Temporal Processing systems
Contacts:
sprugnoli[AT]fbk.eu
Partners:
- Tommaso Caselli, Computational Lexicology & Terminology Lab, VU Amsterdam.
Unit role:
- Development of annotation guidelines
- Manual annotation
- Co-organization of the EVENTI evaluation exercise
