Learning First-person Knowledge through Interactions: towards Effective Design Tools

This paper introduces an approach that enables a design tool to learn first-person knowledge through interactions. A design tool of this kind embodies learning and adaptive behaviours. The contextual knowledge in using the tool to solve a particular design problem can be captured, made available and adapted to a designer’s decision-making when he or she is confronted with a similar or a new problem at a later time. An effect of this is an improvement in design efficacy, in that a design task can be recognized from interactions and experiences for similar tasks and support can be provided to aid decision-making. The implemented prototype system is applied to assist the use of an optimization tool (the Matlab Optimization Toolbox) in design. The system learns knowledge from how Matlab is utilized in solving various optimization problems and uses the learned concepts to affect the tool’s future use. Experiments evaluate the effectiveness of this prototype system in recognizing optimization problems in various design optimization scenarios.