Öztürel, İsmet Adnan
This thesis presents a novel approach to assigning types to expressive Discourse Representation Structure (DRS) meaning representations. In terms of linguistic analysis, our typing methodology couples together the representation of phenomena at the same level of analysis that was traditionally considered to belong to distinctive layers. In the thesis, we claim that the realisation of sub-lexical, lexical, sentence and discourse-level phenomena (such as tense, word sense, named entity class, thematic role, and rhetorical structure) on the surface can be represented as variations of values that belong to the same typed category within our cross-level typing technique. We show the implications of our approach on the computational modelling of natural language understanding (NLU) using Combinatory Categorial Grammar, specifically in the context of one of the core NLU tasks, semantic parsing. We present that crosslevel type-assigned logical forms deliver compact lexicon representations and help re-formalise search space constraining tasks, such as Supertagging, as part of the semantic analysis, whereas such approaches were only used in syntactic parsing. We empirically demonstrate the effectiveness of using a training objective that is based on masking the typed logical forms in pre-training models to obtain re-usable lexical representations. Our results indicate that improved model performance on parsing open-domain text to DRS is possible when the embedding layer of an encoder-decoder model such as Transformer is initialised with weights that are distilled from a model that is pre-trained using our objective.


Hierarchical behavior categorization using correlation based adaptive resonance theory
Yavaş, Mustafa; Alpaslan, Ferda Nur; Department of Computer Engineering (2011)
This thesis introduces a novel behavior categorization model that can be used for behavior recognition and learning. Correlation Based Adaptive Resonance Theory (CobART) network, which is a kind of self organizing and unsupervised competitive neural network, is developed for this purpose. CobART uses correlation analysis methods for category matching. It has modular and simple architecture. It can be adapted to different categorization tasks by changing the correlation analysis methods used when needed. Cob...
Supertagging with combinatory categorial grammar for dependency parsing
Akkuş, Burak Kerim; Çakıcı, Ruket; Department of Computer Engineering (2014)
Combinatory Categorial Grammar (CCG) categories contain syntactic and semantic information. CCG derivation trees can be used in extracting partial dependency structures by providing the missing information in order to build complete dependency structures. Therefore, CCG categories are sometimes referred to as supertags. The amount of information encoded in supertags makes it possible to create very accurate and fast parsers as supertagging is considered ``almost parsing''. In this thesis, a maximum entropy ...
Semantic video analysis for surveillance systems
Kardaş, Karani; Coşar, Ahmet; Çiçekli, Fehime Nihan; Department of Computer Engineering (2018)
This thesis presents novel studies about semantic inference of video events. In this respect, a surveillance video analysis system, called SVAS is introduced for surveillance domain, in which semantic rules and the definition of event models can be learned or defined by the user for automatic detection and inference of complex video events. In the scope of SVAS, an event model method named Interval-Based Spatio-Temporal Model (IBSTM) is proposed. SVAS can learn action models and event models without any pre...
A type system for combinatory categorial grammar
Erkan, Güneş; Bozşahin, Hüseyin Cem; Department of Computer Engineering (2003)
This thesis investigates the internal structure and the computational representation of the lexical entries in Combinatory Categorial Grammar (CCG). A restricted form of typed feature structures is proposed for representing CCG categories. This proposal is combined with a constraint-based modality system for basic categories of CCG. We present some linguistic evidence to explain why both a uni cation-based feature system and a constraint-based modality system are needed for a lexicalist framework. An implem...
Machine Learning and Rule-based Approaches to Assertion Classification
Uzuner, Oezlem; Zhang, Xiaoran; Sibanda, Tawanda (Oxford University Press (OUP), 2009-01-01)
Objectives: The authors study two approaches to assertion classification. One of these approaches, Extended NegEx (ENegEx), extends the rule-based NegEx algorithm to cover alter-association assertions; the other, Statistical Assertion Classifier (StAC), presents a machine learning solution to assertion classification.
Citation Formats
İ. A. Öztürel, “CROSS-LEVEL TYPING THE LOGICAL FORM FOR OPEN-DOMAIN SEMANTIC PARSING,” Ph.D. - Doctoral Program, Middle East Technical University, 2022.