LX-TimeAnalyzer is a freely available online service for the extraction of temporal information from Portuguese text. It was developed and is mantained by the NLX-Natural Language and Speech Group at the University of Lisbon, Department of Informatics.
LX-TimeAnalyzer extracts temporal information from Portuguese texts. Given an input text, it finds the following elements:
- Temporal expressions, which are expressions that occur in the input text and that refer to dates and times;
- Events terms, which are words that refer to events that happen or hold at some point in time;
- Temporal relations between these times and events, i.e. the temporal ordering among these entities, according to the input text. The possible relations are IS BEFORE (the first event precedes the second event), IS AFTER (the first event follows the second event), and OVERLAPS (both events overlap in time).
The temporal relations are also extracted in relation to the reference time given.
The output of LX-TimeAnalyzer is presented in two columns.
The input text is shown on the left column. The time expressions that occur in this text are highlighted in green. Hovering the mouse cursor over them shows information about the time or date that they refer to, in a normalized format. Event terms are presented in orange. Hovering the mouse cursor over any of these elements shows the temporal relations which the mentioned entity enters.
The right column shows a temporal graph for the paragraph on its left. In this temporal graph, nodes represent times (including the document's creation time) and events. Green nodes represent times, and orange nodes represent events. Temporal relations are also encoded in this graph.
Temporal precedence is shown with an arrow between two nodes. The arrow's direction indicates the flow of time: an arrow going from a node m to a node n means that the entity represented by m precedes in time the entity represented by n. Most arrows go from left to right, and usually (but not always) reading these graphs in that direction matches the temporal flow of the described events and times.
Entities that temporally overlap are represented in this graph with nodes that tend to be vertically aligned, but this is not always feasible. Therefore, not all entities that temporally overlap are necessarily placed on the same column, and not all nodes that are aligned vertically represent entities that temporally overlap, but many do.
If the text contains more than one paragraph, a graph for the entire text is shown below these two columns.
The work leading to LX-TimeAnalyzer was supported by FCT—Fundação para a Ciência e Tecnologia—under the research grant SFRH/BD/40140/2007.
Irrespective of the most recent version of this tool you may use, when mentioning it, please cite this reference:
- Costa, Francisco and António Branco, 2012, "Aspectual Type and Temporal Relation Classification". In Proceedings, EACL2012 - 13th Conference of the European Chapter of the Association for Computational Linguistics, Avignon, France, 23-27 April 2012, pp.266-275.
- Costa, Francisco and Branco, António. 2012. "Extracting Temporal Information from Portuguese Texts". In Helena Caseli, Aline Villavicencio, António Teixeira e Fernando Perdigão (eds.), Computational Processing of the Portuguese Language-10th International Conference, PROPOR 2012, volume 7243 de Lecture Notes in Artificial Intelligence, pp. 99-105, Berlin: Springer.
- Costa, Francisco and Branco, António. 2012. LX-TimeAnalyzer: A Temporal Information Processing System for Portuguese. Technical Report DI-FCUL-TR-2012-01, Universidade de Lisboa, Faculdade de Ciências, Departamento de Informática.
- Costa, Francisco e Branco, António. 2013. "Temporal Relation Classification Based on Temporal Reasoning". In Proceedings of the 10th International Conference on Computational Semantics, Potsdam.
- Costa, Francisco. 2013. Processing Temporal Information in Unstructured Documents. Doctoral dissertation, Universidade de Lisboa, Lisbon.
Contact us using the following email address: 'nlxgroup' concatenated with 'at' concatenated with 'di.fc.ul.pt'.
LX because LX is the "code" name Lisboners like to use to refer to their hometown.