DSTC Logo spacer
spacer spacer
spacer
spacer
spacer spacer

spacer
Information Ecology
Paradoxically, a consequence of the information explosion is diminishing awareness; disciplines are becoming increasingly specialized, individuals, groups, organizational units are becoming ever more insular. Disparate islands of (community) knowledge are being formed and evolving with little or no inter-connection to other islands. Don Swanson's discovery of a cure for Raynaud's disease with dietary fish oils by serendipitously bridging two distinct islands of medical knowledge suggests that there may be many islands of knowledge that can be bridged with beneficial results.

The lack of awareness impinges on scientists' ability to produce hypotheses seeded by knowledge outside of their own specialization. Similarly, individuals, groups, and units within an organization could benefit greatly from a broader awareness. The underlying reason for lack of awareness ties back to the individual’s bounded cognitive resources: time, information and computational capacity.

Lack of awareness is closely related to a lack of introspection, in other words, the individual does not know that (s)he knows something in a given context. Lack of introspection in an enterprise setting contributes to "organizational ignorance", a consequence of which is, for example, work/tasks having to be redone, or the collection of information that has already been collected.

In summary, human beings have limited cognitive resources. They handle scarce information because that is all the conscious individual can handle. Consequently, their introspective abilities and awareness are impaired. This project attempts to produce knowledge-based technologies that promote awareness and introspection. In a nutshell, this involves developing:
computational forms of socio-cognitive knowledge representation called "semantic spaces"

techniques that mimic/support human reasoning in relation to information

suitable models of context to condition the knowledge processing systems

appropriate frameworks for evaluating the knowledge processing systems
Details of two research initiatives are given below.

Social networks
Investigate the theory and application of knowledge in online communities from a socio-cognitive perspective. Computational models of the semantics of electronic communications provide the basis of techniques which discover the explicit and tacit knowledge in the communication between people. This allows networks of people/information to be constructed based on semantic context. In addition, tacit knowledge aids those inside the community, and others outside, to have a better understanding of the community in all its facets.

Information Inference
Research in developing computational devices which draw inferences from textual information which correlate with human inference. Computational models from cognitive science are used to represent the meaning of word, or word compounds in a semantic space. Inference is driven by computations within the space, for example, a high degree of information flow between two words equates with a strong inferential connection (either explicit or implicit). One of the primary goals of this research is to mimic a form of human reasoning called abduction. Abduction is a process which can generate new knowledge via hypothesis formation. Automated abduction will be applied to knowledge discovery in medical literature and the composition of web services.

Contact Point:
Dr Peter Bruza
Phone +61 7 3365 4310
Email

Staff:
Professor Peter Bruza DSTC UQ
Rob McArthur DSTC UQ
Dr Dawei Song DSTC UQ
Dr Richard Cole ITEE/DSTC UQ
Zeeniya Bari ITEE/DSTC UQ (PhD Student)

Associate Researchers:
(ITEE/UQ)
(ITEE/UQ)
(ITEE/UQ)
(Centre for Information Technology Innovation, QUT)
, University of Glasgow, UK.
, University of Nijmegen, the Netherland
, The Chinese University of Hong Kong
, (University of Paisley, PhD student)
Helen Huang (ITEE/UQ, PhD student)
Xin Yan (ITEE/UQ, PhD student)
James Cole (ITEE/UQ, PhD student)
spacer
img
spacer
spacer