Event and Temporal Relation Extraction from Croatian Newspaper Texts
نویسندگان
چکیده
Event extraction and temporal relation extraction are the subjects of extensive research, which has been additionally motivated by focused evaluation exercises such as TempEval. In this paper we present the work on supervised event and temporal relation extraction from Croatian newspaper texts. Taking into account the limited availability of linguistic tools for Croatian, we focus our research around simple lexical features. We manually annotated a newspaper corpus or events and temporal relations in Croatian according to the TimeML and TimeBank guidelines. Experimental evaluation yielded promising results: F1 scores of up to 77% for event identification, 48% for event classification, and 51% for temporal relation classification. Luščenje dogodkov in časovnih relacij iz hrvaških časopisnih besedil Luščenje dogodkov in časovnih relacij je zelo živahno raziskovalno področje, ki se je še posebej razmahnilo s pojavom skupinskih evalvacijskih pobud, kot je na primer TempEval. V pričujočem prispevku predstavljamo sistem za nadzorovano luščenje dogodkov in časovnih relacij iz hrvaških časopisnih besedil. Glede na to, da je dostopnost jezikovnih orodij za hrvaščino omejena, se v raziskavi osredotočamo zgolj na enostavne leksikalne lastnosti. Pri ročnem označevanju dogodkov in časovnih relacij v korpusu časopisnih besedil smo uporabljali smernice TimeML in TimeBank. Eksperimentalno vrednotenje rezultatov je zelo spodbudno, saj za prepoznavanje dogodkov F1 znaša 77%, za klasifikacijo dogodkov 48% in za klasifikacijo časovnih relacij 51%.
منابع مشابه
GPKEX: Genetically Programmed Keyphrase Extraction from Croatian Texts
We describe GPKEX, a keyphrase extraction method based on genetic programming. We represent keyphrase scoring measures as syntax trees and evolve them to produce rankings for keyphrase candidates extracted from text. We apply and evaluate GPKEX on Croatian newspaper articles. We show that GPKEX can evolve simple and interpretable keyphrase scoring measures that perform comparably to more comple...
متن کامل05151 Summary - Annotating, Extracting and Reasoning about Time and Events
Newspaper articles and other natural-language texts describe actions, events, and states of affairs. A crucial first step toward the automatic extraction of information from these texts—for use in such applications as automatic question answering or summarization—is the capacity to identify what events are being described and to make explicit when these events occurred and which temporal relati...
متن کاملنقش حروف ربط زمان دار در تعیین رابطۀ زمانی بین رویدادهای فعلی در پیکرۀ متون زبان فارسی معاصر
This paper involves in prediction of temporal relation between tensed-verb events on the basis of conjunctions in texts. For this purpose, tensed verb event data were extracted from Contemporary Persian Corpus and were examined carefully. The temporal relations between events were identified. After analyzing data on the basis of temporal relation according to Bird’s and Allen’s categorization, ...
متن کاملThe METER Corpus: A corpus for analysing journalistic text reuse
As a part of the METER (MEasuring TExt Reuse) project we have built a new type of comparable corpus consisting of annotated examples of related newspaper texts. Texts in the corpus were manually collected from two main sources: the British Press Association (PA) and nine British national newspapers that subscribe to the PA newswire service. In addition to being structured to support efficient s...
متن کاملUTHealth at SemEval-2016 Task 12: an End-to-End System for Temporal Information Extraction from Clinical Notes
The 2016 Clinical TempEval challenge addresses temporal information extraction from clinical notes. The challenge is composed of six sub-tasks, each of which is to identify: (1) event mention spans, (2) time expression spans, (3) event attributes, (4) time attributes, (5) events’ temporal relations to the document creation times (DocTimeRel), and (6) narrative container relations among events a...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2012