Time is a useful dimension to explore in text databases especially when historical and factual information is concerned. As documents generally refer to different events and time periods, understanding the focus time of key sentences, defined as the time the content refers to, is a crucial task to temporally annotate a document. In this paper, we leverage a bag of linked entities representation of sentences and temporal information from Wikipedia and DBpedia to implement a novel approach to focus time estimation. We evaluate our approach on sample datasets and compare it with a state of the art method, measuring improvements in MRR.
Leveraging linked entities to estimate focus time of short texts
C. Morbidoni;
2018-01-01
Abstract
Time is a useful dimension to explore in text databases especially when historical and factual information is concerned. As documents generally refer to different events and time periods, understanding the focus time of key sentences, defined as the time the content refers to, is a crucial task to temporally annotate a document. In this paper, we leverage a bag of linked entities representation of sentences and temporal information from Wikipedia and DBpedia to implement a novel approach to focus time estimation. We evaluate our approach on sample datasets and compare it with a state of the art method, measuring improvements in MRR.File in questo prodotto:
Non ci sono file associati a questo prodotto.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
