Studying the Usability of Relevant Proximity Ranking,

Abstract

A ranked list returned by an information retrieval system lists the documents in the order they are expected to match the user's query: the first document is most likely to be the most relevant, the second is the next one most likely to be helpful, and so on. It is expected that the user will follow these recommendations, starting at the top of the list and following it down, reading documents one by one. It is a well-known and widely accepted method for presenting the retrieved information and helping the user to find relevant documents. Ideally, the user will see all the relevant documents before any non-relevant ones, though quite often the relevant documents appear to be scattered all over the ranked list.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 1999
Accession Number
ADA365684

Entities

People

  • Anton Leuski

Organizations

  • University of Massachusetts Amherst

Tags

DTIC Thesaurus Topics

  • Algorithms
  • Automatic
  • Clustering
  • Computer Science
  • Embedding
  • Information Retrieval
  • Materials
  • Precision
  • Three Dimensional
  • Two Dimensional
  • Visualizations

Readers

  • Computational Linguistics
  • Information Retrieval
  • Systems Analysis and Design

Technology Areas

  • AI & ML
  • AI & ML - Information Retrieval