My Research

Highly motivated students interested in PhD research in knowledge graphs, natural language processing, graph/network embedding and related areas are welcome to get in touch and apply!

Research interests:

  • Generative & Neuro-symbolic AI
  • knowledge graphs
    • Graph representation learning
    • Knowledge extraction from unstructured data (text, images, etc.)
  • Natural language processing
    • Large language models (LLMs)
    • Complex question answering
    • Language generation from text/knowledge

In the distant past…

  • Semantic Web
    • Knowledge graph construction & representation learning
    • Ontology languages (DAML+OIL, OWL, RDF/RDFS, SWRL), reasoning performance prediction & optimisation
    • SPARQL query processing
    • Data management for bioinformatics research
    • Effective visualisation of complex ontologies
    • Information visualisation
  • Software engineering
    • Software engineering knowledge integration & analysis
    • Software requirements analysis, software complexity
    • Formal design methods (Alloy, Z, PVS)
    • Program analysis

PhD Supervision

Current PhD students

Completed PhD students

  • Bhagya Hettige (2017–2021): The COEUS project: how can the application of predictive analytics to digitalised patient data revolutionise healthcare?, joint supervision with Wray Buntine, Teresa Wang, Suong Le. [ADC’20 (best student paper award), ECAI’20, PAKDD’20]

  • Ying Yang (2016–): Effective Visualisations of Large Biomedical Ontologies and Associations, joint supervision with Michael Wybrow, Tobias Czauderna. [InfoVis’19]

  • Vishwajeet Kumar (2016–2020): Natural language question generation, IITB-Monash Academy, joint supervision with Ganesh Ramakrishnan (IITB). [PAKDD’18, CoNLL’19, EMNLP’19, ISWC’19, AACL-IJCNLP’20]

  • Ayesha Sadiq (2017–2019): Extracting Access Permission-based Specifications from a Sequential Program, joint supervision with Chris Ling, Li Li and Ijaz Ahmed. [ASE’19, JSS, ASE’20]

  • Carlos Oliveira (2017-2019): Mapping the Effectiveness of Automated Test Suite Generation Techniques, joint supervision with Aldeida Aleti. [GECCO’19]

  • Wudhichart Sawangphol (2013–2017): Constraint-based Reasoning for Description Logics with Concrete Domains and Aggregations, joint supervision with Guido Tack.

  • Nader Chmait (2014–2017): Understanding Collective Intelligence in Agent-Based Systems: an Information-Theoretic Approach to the Measurement and Comparison of Intelligence in Groups, joint supervision with David Dowe, David Green. [BBS, AGI’17, ECAI’16]

  • Steve Quenette (2015–2017): Semantics & implementation of computational software families - software development as an inverse process, joint supervision with David Abramson, Heinz Schmidt.

  • Chetana Gavankar (2015–2017): Ontology population and enriching search across ontologies: methodologies and challenges, IITB-Monash Academy, joint supervision with Ganesh Ramakrishnan (IITB). [K-Cap’15, ISWC’16]

Research Grants

  • Visualisation of large, complex networks through small, beautiful diagrams, ARC Discovery Project, 2014–2016.

  • Skeletome - A Curated Online Knowledge Base Integrating Clinical and Biological Information on Skeletal Dysplasias. ARC Linkage Project, 2010–2012.

Professional Activities

Program co-chair

  • 6th Joint International Conference on Semantic Technology (JIST 2016).

  • 20th International Conference on Engineering of Complex Computer Systems (ICECCS 2015).

  • ACM/IEEE International Joint Conference on Digital Libraries 2010 (JCDL 2010).

Conference organisation

  • JIST 2017.

  • ISWC 2013.

  • JCDL 2010 & ICADL 2010.

Journal Reviewer

  • Advances in Software Engineering (ASE)

  • IEEE Transactions on Software Engineering (TSE)

  • International Journal on Knowledge and Information Systems (KAIS)

  • Journal of Software (JSW)

  • Science of Computer Programming (SCP)

Conference Reviewer

  • AAAI, APSEC, ACSC, ASWEC, FM, ICECCS, ICFEM, K-CAP, IJCAI, ISWC, WWW, …

Some Past Projects

I have participated in a number of research projects while I was at the eResearch Lab, UQ. Below you can find some information about these projects.

Skeletome
A Curated Online Knowledge Base for Integrating Clinical and Biological Information, ARC Linkage Project, (2010–2012).

Skeletome is a Semantic Web-enabled knowledge base for skeletal dysplasias; it aims at utilising novel Semantic Web and Web 2.0 technologies to create a community-driven, expert-curated semantic knowledge base that supports semantic content annotation, aggregation and visualisation.

PODD
Phenomics Ontology Driven Data repository, NeAT (2009–2011).

Co-funded by Australian National Data Service (ANDS) and Australian Research Collaboration Service (ARCS) under the National e-Research Architecture Taskforce (NeAT), the PODD (Phenomics Ontology Driven Database) project. PODD is a multi-disciplinary, collaborative, open-source project aimed at developing a state-of-the-art online data repository to meet the data management needs of the Australian phenomics research community. Being world’s first ontology-driven data repository, PODD is designed to be more extensible, flexible and open, enabling effective data storage, contextualisation, sharing, management and publishing of raw data. Responsibilities include (1) overall ontology-driven system architecture modelling and design; (2) developing the PODD ontology in the OWL ontology language; and (3) leading all aspects of the software development activities, including the development of the entire backend system.

BioMANTA
Modelling and Analysis of Biological Network Activity, Pfizer, (2007–2009).

The BioMANTA project focusses on the computational modelling and analysis, primarily using Semantic Web technologies, of large-scale protein-protein interaction and compound activity networks across a wide variety of species. A range of information such as kinetic activity, tissue expression, sub-cellular localisation and disease state attributes is included in the resulting data model. Semantic Web technologies allow for flexible data integration, inherent inferencing capabilities and advanced machine learning methods. The BioMANTA project aims to bring together these areas through the modelling of PPI data in a semantic web framework, using technologies such as RDF and OWL, and then to apply in silico and experimental analysis methods to analyse aspects of the BioMANTA data model.