Application areas


Currently the DIT AI Group has a range of active projects. These are listed below with links to further information.

Tackling Concept Drift Without Labelled Instances

When creating a model using machine learning the availability of labelled instances is vital. Once the model has been built their importance decreases as long as the concept remains static. However if the concept changes labelled instances are once again imperative. The focus of this research is to minimise the need for labelled instances when tackling concept drift.
More info

CBTV: Case Base Topology Visualisation

A suite of visualisation tools is being developed to be used to aide machine learning tasks through visualisations. Machine learning tasks addressed include similarity measure design and selection, active learning and case base maintenance.
More info

Analysing Dail Questions

The availability of the full text of Irish parliamentary questions offers opportunities for using machine learning techniques to examine the currently much discussed role of elected representatives (TDs) in the Irish parliamentary system (the Dail). This project is investigating the labelling of Dail questions using Active Learning techniques.
More info

LOK8

The LOK8 (pronounced locate) project is a joint project between the School of Computing and the Digital Media Centre which is investigating personalised location based services using using mobile devices ranging from auditory user interfaces on Bluetooth earpieces to intelligent avatars equipped with cameras, microphones, speakers, and projectors.
More info

Comenius 2.0

Comenius 2.0 is a translational research project developing novel technology for English-as-a-second-language teaching that identifies the difficulty level of online articles and recognised grammatical patterns and words of interest in these articles. This information is used as the basis for a search engine and a suite of exercise generation tools.
More info

Using SLAM & RFID for Domestic Robots

Novel locationing technologies, such as RFID, offer new potentials for localisation and mapping for robotics applications. This project investigates how these new technologies can be applied to the task of simultaneous localisation and mapping.
More info

CBM for Text Classification

Case-base editing/Maintenance (CBM) is an important part of the case-based reasoning (CBR) process and is important for maintaining the quality of CBR classifiers. This project is investigating the use of novel case base editing techniques with particular focus on text classification.
More info

Intelligent agent design tools for serious games

This project is developing design tools for use in the creation of intelligent agents to act as background characters for serious games.
More info

Credit Scoring Using Machine Learning

For financial institutions and the economy at large, the role of credit scoring in lending decisions cannot be overemphasised. An accurate and well-performing credit scorecard allows lenders to control their risk exposure through the selective allocation of credit based on the statistical analysis of historical customer data. This work identifies and investigates a number of specific challenges that occur during the development of credit scorecards.

Resources

Datasets:

Software: