Automatic detection and recognition of hazardous chemical agents

The number and use of hazardous chemical compounds are increasing, providing an important and critical application area of detector devices. In addition to the devices, also extremely reliable detection algorithms must be implemented. The design of such algorithms has traditionally been an analytical process demanding a vast amount of work and expertise. Thus, there is a strong interest of automatic machine learning methods. In this study, several machine learning methods are applied to a detector device measuring the ion mobility distribution for detecting and recognizing chemical warfare agents. The experimental results indicate that one of the proposed methods, the Bayesian classifier based method, is applicable even for critical applications.