Output details
11 - Computer Science and Informatics
University of Strathclyde
Article title
Evaluating epistemic uncertainty under incomplete assessments
Type
D - Journal article
Title of journal
Information Processing and Management
Article number
-
Volume number
44
Issue number
2
First page of article
811
ISSN of journal
0306-4573
Year of publication
2008
Number of additional authors
2
Additional information
<17>Introducing the fundamental new concept of epistemic uncertainty to Information Retrieval evaluation, this paper provides a method for estimating when we have enough confidence that system comparisons are fair and metrics for estimating this confidence. Therefore we can tell for the first time when there may be too much uncertainty in an evaluation to make confident judgments that one system is better than another. This work, being based on the dominant model of IR evaluation and applicable to all published work in IR that use this evaluation model, has the capacity to be a major influence in the field.
Interdisciplinary
-
Cross-referral requested
36 - Communication, Cultural and Media Studies, Library and Information Management
Research group
B - Information Systems
Citation count
5
Proposed double-weighted
No
Double-weighted statement
-
Reserve for a double-weighted output
No
Non-English
No
English abstract
-
