Evolutionary learning of graph layout constraints from examples

摘要: We propose a new evolutionary method of extracting user preferences from examples shown to an automatic graph layout system. Using stochastic methods such as simulated annealing and genetic algorithms, systems can find good using evaluation function which calculate how given is. However, the is usually not known beforehand, it might vary user. In our system, users show system several pairs bad examples, infers programming technique. After evolves reflect user, used general for laying out graphs. The same technique be wide range adaptive interface systems.