Output details
11 - Computer Science and Informatics
Robert Gordon University
Return to search
Output 0 of 0 in the submission
Output title
Applying Machine Translation Evaluation Techniques to Textual CBR
Type
E - Conference contribution
Name of conference/published proceedings
Case-Based Reasoning Research and Development: Proceedings of the 18th International Conference on Case-Based Reasoning, ICCBR 2010 (LNCS Volume 6176)
Volume number
6176
Issue number
-
First page of article
21
ISSN of proceedings
1611-3349
Year of publication
2010
URL
-
Number of additional authors
3
Additional information
In the textual case-based reasoning (TCBR) approach, solutions in the form of unstructured text are difficult to evaluate correctly. Variations in word choice, errors and abbreviations can mask underlying similarity. Worse, varying descriptions of the same event can lead to low similarity scores for good solutions.
Similar problems in machine translation have been addressed by evaluating against multiple human translations. This paper adapts these techniques to evaluation in TCBR.
Comparison with human similarity judgements show this approach to improve evaluation of textual solutions.
Interdisciplinary
-
Cross-referral requested
-
Research group
None
Citation count
1
Proposed double-weighted
No
Double-weighted statement
-
Reserve for a double-weighted output
No
Non-English
No
English abstract
-