A Bayesian Network Approach to Mode Detection for Interactive Maps

This paper describes a mode detection system for online pen input that employs a Bayesian network to combine classification results and context information. Previous monolithic classifiers were not able to provide sufficient performance to be used in the domain of crisis management, where robust interaction is extremely important. To enhance mode detection for the intended target domain of crisis management, domain specific pen gesture data was used to train the four different classifiers and to calculate the conditional probabilities used in the Bayesian network. Mode detection, which is used to distinguish between different types of pen input such as deictic gestures, handwritten text, and iconic objects, clearly profited from this new approach. The error rate dropped from 9.3% for a monolithic system to 4.0% for the new mode detection system.