The Natural Language Processing Research Group (NLPRG) brings together the fields of Computer Science, Artificial Intelligence, and Computational Linguistics. This field is sometimes also referred to as Human Language Technologies. We are particularly concerned with language interfaces between man and machine and look at ways of how computers can understand, process and generate human language. There are several components that fall under this research group, including:
Extracting Information from Textual Resources
Textual data contains a lot of information that humans need to sieve through to find the relevant and required information. NLP techniques can be used to extract information that is specific to the user’s requirements. This can also be applied to specific domains, such as laws and legal contracts.
There are several areas where Maltese-English machine translation could be applied in, especially in domain-specific scenarios, such as health (hospital scenario to facilitate patient-doctor communication). We are also interested in using Maltese as a pivot language for translating from Arabic to English, and preliminary experiments show that this is a very viable approach.
Speech recognition for Maltese is one of our research priorities. We are currently annotating speech files, with the aim that these can eventually be used to apply machine learning techniques and train an automatic speech recognition system.
Development of Maltese Language Resources
Finalisation of a morphological analyser, further development of syntax analysis, the creation of a named entity recognizer. Maltese can only continue to develop in the digital era if the necessary linguistic processing tools are available.
NLP for Language Learning
Use Natural Language Processing techniques to facilitate language learning – both first language learners (children) and second language learners (children and adults). Our main interest in this aspect is to create games that could assist students to learn and further develop their knowledge in a particular language.
Generation of Text from Images
Image analysis using Deep learning to generate textual representations of those images. Particularly, at the moment we are focusing on face images and using machine learning techniques to produce textual descriptions of those images.
For more information, contact Dr Claudia Borg.