Abstract: An information search trail recommendation method based on the Markov chain model and case-based reasoning is proposed. A laboratory user experiment was designed to evaluate the proposed method. The experimental results demonstrated that novice searchers have a positive attitude toward the search trail recommendation and a willingness to use the recommendation. Importantly, this study found that the search trail recommendation could effectively improve novice searchers’ search performance. This finding is mainly reflected in the diversity of information sources and the integrity of the information content of the search results. The proposed search trail recommendation method extends the application scope of information recommendations and provides insights to improve the organization and management of online information resources.

Keywords: Web search, search process, search trail recommendation, Markov chain, case-based reasoning