We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations

DBLP - CS Bibliography


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Information Retrieval

Title: Atomized Search Length: Beyond User Models

Abstract: We argue that current IR metrics, modeled on optimizing user experience, measure too narrow a portion of the IR space. If IR systems are weak, these metrics undersample or completely filter out the deeper documents that need improvement. If IR systems are relatively strong, these metrics undersample deeper relevant documents that could underpin even stronger IR systems, ones that could present content from tens or hundreds of relevant documents in a user-digestible hierarchy or text summary. We reanalyze over 70 TREC tracks from the past 28 years, showing that roughly half undersample top ranked documents and nearly all undersample tail documents. We show that in the 2020 Deep Learning tracks, neural systems were actually near-optimal at top-ranked documents, compared to only modest gains over BM25 on tail documents. Our analysis is based on a simple new systems-oriented metric, 'atomized search length', which is capable of accurately and evenly measuring all relevant documents at any depth.
Comments: 13 pages, 6 figures
Subjects: Information Retrieval (cs.IR); Computation and Language (cs.CL)
Cite as: arXiv:2201.01745 [cs.IR]
  (or arXiv:2201.01745v1 [cs.IR] for this version)

Submission history

From: John Alex [view email]
[v1] Wed, 5 Jan 2022 18:10:30 GMT (81kb,D)

Link back to: arXiv, form interface, contact.