For her research Bogaard used data from the National Library of the Netherlands. The study revealed that analyzing search logs with predefined metadata values can uncover distinct search patterns within different sections of the collection. These insights allow for concrete recommendations to enhance the search system and collection management. Bogaard further employed a clustering algorithm to group sessions based on metadata, unveiling specific user needs across various parts of the collection.
The findings indicate that leveraging metadata can provide valuable insights into how users conduct searches within different areas of a digital library. Consequently, it offers recommendations to enhance access to the collection, while striking a balance between privacy concerns and the information required to support users effectively. The study's results were presented through MAGUS, a novel session visualization tool that combines graphical representations of search behavior with color-coded metadata visualization. MAGUS has proven effective in identifying users' areas of interest within the collection and discerning different search behavior types.
More info on the thesis ‘Using metadata to understand search behavior in digital libraries’.