My Research

I am affiliated with the INSIGHT Centre for Data Analytics since it was established in 2014, first as a Research Fellow in NUI Galway and now as a Funded Investigator at DCU. Since 2018 I am also a Funded Investigator in the I-Form Advanced Manufacturing Research Centre at DCU.

I have recently developed a research programme named Knowledge-Informed Deep Representation Learning, aiming at exploring the possibility of building better deep representations by exploring and exploiting computational models and neural models and their synergies. One of the hypotheses is that we can learn better deep representations by extracting knowledge from Deep Architectures and then use such knowledge to combine model-driven and data-driven learning, either injecting it back into the network or analysing it. Advances in this area would not only improve explainability of deep learning models, but they would also help dealing with non-stationary data distributions or/and situations in which there is very little data available to fully train a Deep Network, as well as handling data bias, understanding how to debug or engineer Deep Architectures and improve interpretability and explainability of decision processes. Numerous challenges still need to be tackled in this area.

In 2018 I have received both industry and research sponsorships to work in this area. Currently I am supervising the following PhD projects:

  • explainable deep learning through graph analysis
  • network-agnostic knowledge transfer in deep learning
  • continuous understanding of additive manufacturing processes and fault detection using deep learning
In the past few years I have also worked on dynamic decision making and adaptive model building. In particular, I have been investigating scalable and interoperable ways of transforming (structured and unstructured) data streams into actionable knowledge. I have been interested in Stream Reasoning for quite a while, targeting two main challenges:
  • hybrid mechanisms to deal with uncertainty through the combination of probabilistic (inductive) and inductive reasoning, applied to incremental rule learning
  • distributed scalable processing of streams investigating formal methods for distributed reasoning

Key application domains I have been working on include but are not limited to: smart cities and smart communities, digital forensics, behaviour analysis and health monitoring, advanced manufacturing, sustainable IT and smart enterprise. My expertise in related areas includes, among others:

  • Knowledge Representation and Reasoning
  • Graph Analytics
  • Deep Representation Learning
  • Semantic Technologies
  • Internet of Things
  • Stream Reasoning and distributed processing
  • Probabilistic inference
  • Complex Event Processing

For a list of publications please check Research gate and DBLP.

For recent updates on activities and outputs within the INSIGHT Centre for Data Analyticis please check my INSIGHT webpage.

For recent updates on activities and outputs within the I-Form Advanced Manufacturing Research Centre, please check the I-Form website.

Current European Projects

Cost Action on Digital Forensics: evidence analysis via intelligent systems and practices.

Past European Projects

CityPulse: Real-Time IoT Stream Processing and Large-scale Data Analytics for Smart City Applications.

EasyReach: Active and Assisted Living.