Learning strategies for personal agents to assist professional users in searching the Web

点击翻译按钮获取中文摘要

Many professional groups have large unmet information needs, with significant consequences for the public. We will take a novel strategy-based approach to the design of agents that assist humans in information search tasks, which represents a significant departure from the dominant paradigm of content-based retrieval. Our goal is to develop automated methods to learn task specific search strategies, embed them in agents, and use them to guide human-agent interactions and search behaviours. To discover the effectiveness of our approach we will evaluate our methods with health care professionals.