Please use this identifier to cite or link to this item:
Title: Detection Of Causlity Relations In Plain Text With The Use Of Word Embeddings
Authors: Γρηγόριος Κ. Μπάστας
Σταφυλοπάτης Ανδρέας-Γεώργιος
Keywords: causality relations
word embeddings
natural language processing
machine learning
Issue Date: 9-Nov-2017
Abstract: Causality detection is one of the most challenging topics in Natural Language Processing (NLP). In this project we tried to cope with this open problem by employing training methods focused on the creation of vector representations of french words. While we only worked on the problem of causality detection in the French language, our methodology is applicable in many other cases thanks to its generality. Our whole project can be separated into three major tasks.The first task pertains to the creation of our training data through the automatic extraction of cause-effect tuples from a syntactically annotated French corpus. For this purpose, we collected non-ambiguous lexical units from the ASFALDA French FrameNet, that denote causality relations. We, therefore, extracted tuples of meaningful sets of wordsthat represent either the cause or the effect of the captured frame. To achieve all of this, we took advantage of the dependency tree of each sentence and the part-of-speech tag of each word.The second task deals with the computational processing of our training data extracted in the previous task, in order to create causal word embeddings based on cause-effectcontext similarity. At this stage, the cause-effect tuples created in the first task are treated in an innovative manner as the training data set for the models Word2vec, SVDand NMF, in such a way as to create causal embeddings.The third task is about the evaluation of our models. We compared the causal proximity of cause-effect test pairs by employing our word embeddings. For the evaluation, weuse the SemEval Task8 test data (partially translated in French).
Appears in Collections:Διπλωματικές Εργασίες - Theses

Files in This Item:
File SizeFormat 
DT2017-0296.pdf1.4 MBAdobe PDFView/Open

Items in Artemis are protected by copyright, with all rights reserved, unless otherwise indicated.