Article Structure
Abstract
This paper presents a novel framework called error case frames for correcting preposition errors.
Introduction
This paper presents a novel framework for correcting preposition errors.
Error Case Frame
An error case frame consists of a verb, cases, and a feedback message as shown in Fig.
Generating Error Case Frames
The method proposed here exploits two sources of corpus data: native and learner corpora.
Correcting Preposition Errors
Preposition errors are corrected by applying the generated error case frames to the target text.
Evaluation
We evaluated the proposed method from two points of View: correction performance and usefulness of feedback messages.
Discussion
As the experimental results show, the proposed method achieves a comparable correction performance with the classifier-/SMT-based methods.
Conclusions
This paper presented a novel framework called error case frames for correcting preposition errors with feedback messages.
Topics
F-measure
- We measured correction performance by recall, precision, and F-measure .Page 7, “Evaluation”
- The simple error case frame-based method achieves an F-measure of 0.189.Page 7, “Evaluation”
- The hybrid methods achieve the best performances in F-measure .Page 7, “Evaluation”
- Table 2: Correction performance in recall (R), precision (P), and F-measure (F).Page 8, “Evaluation”
See all papers in Proc. ACL 2014 that mention F-measure.
See all papers in Proc. ACL that mention F-measure.
Back to top.
