Structure vs. content in hierarchical corpora

摘要

We propose and describe Parameterized filesystem HITS (PFH), a lightweight rearrangement algorithm that performs a re-ranking of existing content-only search results in tree-like hierarchical corpora, such as a filesystem. PFH does this by combining the content analysis of the results with structural analysis of how they are organized. An a parameter is used to vary the structure-content bias, allowing us to observe changes in performance when we place more emphasis on structure or on content. Using real and simulated data, experiments are provided to show that a = 0.8 (i.e, 20% structure and 80% content) can substantially boost Mean Reciprocal Rank while keeping other IR metrics steady. Such an algorithm may be useful for building retrieval systems over hierarchical content such as Desktop Search, technical documents and books.