A5056593729- Natural Language Processing Techniques
- Topic Modeling
- Semantic Web and Ontologies
- Multimodal Machine Learning Applications
- Sentiment Analysis and Opinion Mining
- Hate Speech and Cyberbullying Detection
- Speech and dialogue systems
- Mental Health via Writing
- Speech Recognition and Synthesis
- Humor Studies and Applications
- Translation Studies and Practices
- Text Readability and Simplification
- linguistics and terminology studies
- Advanced Text Analysis Techniques
- Digital Mental Health Interventions
- Misinformation and Its Impacts
- Biomedical Text Mining and Ontologies
- Mental Health Research Topics
- Machine Learning in Healthcare
- Advanced Graph Neural Networks
- Intelligent Tutoring Systems and Adaptive Learning
- Text and Document Classification Technologies
- Advanced Malware Detection Techniques
- Child and Adolescent Psychosocial and Emotional Development
- Lexicography and Language Studies
Ollscoil na Gaillimhe – University of Galway
2011-2023
Insight (China)
2020-2022
Linköping University
2019
Stockholm University
2019
National University of Ireland
2016
Enterprise Ireland
2012
Under-resourced languages are a significant challenge for statistical approaches to machine translation, and recently it has been shown that the usage of training data from closely-related can improve translation quality these languages. While within same language family share many properties, under-resourced written in their own native script, which makes taking advantage similarities difficult. In this paper, we propose alleviate problem different scripts by transcribing script into common...
Abstract Machine translation is one of the applications natural language processing which has been explored in different languages. Recently researchers started paying attention towards machine for resource-poor languages and closely related A widespread underlying problem these systems linguistic difference variation orthographic conventions causes many issues to traditional approaches. Two written two orthographies are not easily comparable but information can also be used improve system....
Recently, there is an increasing tendency to embed functionalities for recognizing emotions from user-generated media content in automated systems such as call-centre operations, recommendations, and assistive technologies, providing richer more informative user profiles. However, date, adding these was a tedious, costly, time-consuming effort, requiring identification integration of diverse tools with interfaces required by the use case at hand. The MixedEmotions Toolbox leverages need...
While neural networks have led to substantial progress in machine translation, their success depends heavily on large amounts of training data. However, parallel corpora are not always readily available. Moreover, out-of-vocabulary words---mostly entities and terminological expressions---pose a difficult challenge Neural Machine Translation systems. Recent efforts tried alleviate the data sparsity problem by augmenting using different strategies, such as external knowledge injection. In this...
Mental health challenges pose considerable global burdens on individuals and communities. Recent data indicates that more than 20% of adults may encounter at least one mental disorder in their lifetime. On the hand, advancements large language models have facilitated diverse applications, yet a significant research gap persists understanding enhancing potential within domain health. other across various an outstanding question involves capacity to comprehend expressions human conditions...
Abstract This work focuses on the extraction and integration of automatically aligned bilingual terminology into a Statistical Machine Translation (SMT) system in Computer Aided scenario. We evaluate proposed framework that, taking as input small set parallel documents, gathers domain-specific terms injects them an SMT to enhance translation quality. Therefore, we investigate several strategies extract align across languages integrate it system. compare two injection methods that can be...
Conversational recommender systems focus on the task of suggesting products to users based conversation flow. Recently, use external knowledge in form graphs has shown improve performance recommendation and dialogue systems. Information from aids enriching those by providing additional information such as closely related textual descriptions items. However, are incomplete since they do not contain all factual present web. Furthermore, when working a specific domain, its entirety contribute...
Mihael Arcan, Marco Turchi, Paul Buitelaar. Proceedings of the 53rd Annual Meeting Association for Computational Linguistics and 7th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2015.
While neural networks have been used extensively to make substantial progress in the machine translation task, they are known for being heavily dependent on availability of large amounts training data. Recent efforts tried alleviate data sparsity problem by augmenting using different strategies, such as back-translation. Along with scarcity, out-of-vocabulary words, mostly entities and terminological expressions, pose a difficult challenge Neural Machine Translation systems. In this paper,...
The task of causal question answering aims to reason about causes and effects over a provided real or hypothetical premise. Recent approaches have converged on using transformer-based language models solve tasks. However, pretrained often struggle when external knowledge is not present in the premise additional context required answer question. To best our knowledge, no prior work has explored efficacy augmenting with for multiple-choice answering. In this paper, we novel strategies...
The automatic translation of domain-specific documents is often a hard task for generic Statistical Machine Translation (SMT) systems, which are not able to correctly translate the large number terms encountered in text. In this paper, we address problems identification bilingual terminology using Wikipedia as lexical resource, and its integration into an SMT system. correct equivalent disambiguated term identified monolingual text obtained by taking advantage multilingual versions...
This publication has emanated from research supported in part by a grant Science Foundation Ireland (SFI) under Grant Number SFI/12/RC/2289, co-funded the European Regional Development Fund, and Enterprise (EI) Innovation Partnership Programme under grant agreement No IP20180729, NURS – Neural Machine Translation for Under-Resourced Scenarios