|
Title:
|
AUTOMATIC IDENTIFICATION OF TIME AND SPACE CATEGORIES IN THE NATURAL LANGUAGE TEXT |
|
Author(s):
|
Alexander Dmitriev, Alla Zaboleeva-Zotova, Julia Orlova, Vladimir Rozaliev |
|
ISBN:
|
978-989-8533-20-3 |
|
Editors:
|
Hans Weghorn |
|
Year:
|
2013 |
|
Edition:
|
Single |
|
Keywords:
|
Text mining; natural language processing; heterogeneous semantic network; topo-temporal logics; Markov-logic networks. |
|
Type:
|
Short Paper |
|
First Page:
|
187 |
|
Last Page:
|
190 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
The paper presents a mining methodology, which is based on processing of spatial and temporal information in natural language text. It is based on methodology capable of recognizing the relationship between events and objects in the text, using spatial and temporal relationships between these events to enhance the semantic coherence of natural language (NL) text. Heterogeneous semantic network with extended family relationships is used to implement relational-situational analysis between the words in the sentence. For correlation lining between spatial and temporal categories and events in the text the mechanism of Markov-logic networks (MLNs) is used, allowing to operate in the peaks of MLNs with the normal logical formulas representing the combined spatial-temporal logic. |
|
|
|
|
|
|
