Investigation of information exploitation techniques for decision-support agents

Abstract

Today, humans are generating vast amounts of data that need to be stored, organized, understood, and prioritized in order to accomplish tasks that are dependent on decision-making. In some settings, such as medical, emergency, and military, it is crucial that the way information is perceived and understood is done accurately and in a timely-manner so that important, sometimes life-changing, decisions are made. With the advances of data science and machine learning, it is possible to create, train, and utilize decision-support agents to help assist humans with the handling and processing of information and data. This thesis examines how the general population perceives, interprets, and values information to model how decision support agents should value and prioritize information.