Research Interests: Anet Potgieter
Adaptive agents are the basic building blocks of a complex adaptive system. These agents act together, interact with each other and the environment, and collectively adapt to changing environmental conditions. This technology is poised to transform the way we model the enterprise and build information systems.
My research interests include the development of simple agents that achieve adaptivity through collective behaviour. These agents form agencies that are distributed in open environments such as the Internet to perform complex intelligent tasks. Each agent can perform only simple operations. When joined together in agencies, intelligence is achieved. I would like to invite MSc and PhD students to join research, using the above technology, in any related area of their choice. Example research areas include:
Complex Adaptive Enterprises The complex adaptive enterprise can learn from and react to global events faster than the competition because of its self awareness and its ability to adapt. An enterprise is self-aware if it understands the interdependencies between its own resources and how these resources and interdependencies contribute to its own competitive advantage. The complex adaptive enterprise then uses this understanding to adapt its business processes and strategies in order to sustain its competitive advantage. In this project, we will research how simple software agents can be integrated into
the enterprise in order to enable the enterprise to function as a complex adaptive enterprise.
Adaptive Value Networks Most leading companies are rapidly trying to develop the ability to leverage scale, outsource key functions and to make products to customer demand, in the most flexible way possible. Current software offerings do not integrate process, value management and intelligence in a way that can exploit opportunities within markets, across business functions, and between businesses in a value network, and cannot fully enable these new ways of doing business. This project will allow a unique combination of adaptive agents, cost and value management, and collaboration across processes
that will drive significant, sustainable competitive advantage.
Adaptive Knowledge Management Knowledge management is concerned with issues involved with identifying, collecting, storing, evaluating, indexing, structuring, extracting, and presenting knowledge used to improve an organization’s productivity. This project will research how adaptive agents can contribute to adaptive knowledge management by collecting knowledge and presenting knowledge in a just-in-time fashion.
Machine Intelligence Machine Intelligence, a subfield of Artificial Intelligence (AI), is concerned with the study and the development of beautiful algorithms to mimic nature for problem solving. Machine learning is an important aspect of this research which involves finding useful models and statistically meaningful patterns out of massive datasets. Emphasis in machine intelligence is on computational intelligence, inductive machine learning, Bayesian learning and reasoning, data mining, and neural networks. Some applications of this research include stock and financial market analysis, anomaly detection, medical diagnosis, bioinformatics and object (speech, face, or handwriting) recognition.
Emergence in Command and Control Applications
Bursaries Available from CSIR Electronic Warfare and Radar Division.
Modelling Emergence in Ecosystems
Bursaries Available from CSIR StelIenbosch.
Agent-based SensorWeb Applications
Anet is a member of the SensorWeb Alliance http://www.sensorweb-alliance.org and has funding available from industry for research in this area.