We take a staged approach to development and adoption.
- Define the problem we’re aiming to solve, the value proposition and our relevant metrics for success for both a proof of concept and a production system built on top of the AI.
- Build and evaluate a proof of concept model on sample data, test the water, see how well we can get it to perform and how effectively we can come up with strategies to improve performance. We assess our performance after each iteration and either accept the performance as adequate and move to the next stage, go through another round of attempted improvements or abandon the project
- Build a platform around the AI to allow the model to deliver business value, generally by enabling client to provide the model inputs and extract and use the outputs.
- Use real-world data and usage to refine the models over time and shape the workflow to maximise adoption by users.
AI enablement capability
- Computer vision
- Data management and analysis
- Expert systems
- Machine learning
- Natural language processing
- Cross industry
- Accommodation and food services
- Administrative and support
- Agriculture, forestry and fishery
- Arts and recreation
- Education and training
- Electricity, gas, water, waste services
- Financial and insurance
- Healthcare and social assistance
- Information, media and telecommunications
- Professional scientific technical services
- Public administration and safety
- Rental, hiring and real estate services
- Retail trade
- Transport, postal and warehousing
- Wholesale trade
Any information included herewith is in no way an endorsement of the listed organisations by the National Artificial Intelligence Centre or CSIRO.