Differences
This shows you the differences between two versions of the page.
Both sides previous revision Previous revision Next revision | Previous revision Next revision Both sides next revision | ||
courses:rg:2012:encouraging-consistent-translation-bushra [2012/10/23 15:33] jawaid |
courses:rg:2012:encouraging-consistent-translation-bushra [2012/10/23 15:49] jawaid |
||
---|---|---|---|
Line 12: | Line 12: | ||
====Approach: | ====Approach: | ||
- | - The core idea of maintaining translation consistencies (TC) is implemented by intrdocuing bias towards TC in form of " | + | - The core idea of maintaining translation consistencies (TC) is implemented by intrdocuing bias towards TC in the form of " |
- BM25 which is used as term weighting function is a well known ranking function in the filed of information retrieval and a refined version of TF-IDF (another ranking function uses in IR). | - BM25 which is used as term weighting function is a well known ranking function in the filed of information retrieval and a refined version of TF-IDF (another ranking function uses in IR). | ||
- | - C< | + | |
- | - C< | + | |
- | - C< | + | - C< |
+ | - C< | ||
Evaluation: | Evaluation: | ||
- Cdec's implementation of Hierarchical MT is used in this work. As we know, hierarchical decoding is also implemented in other MT systems such as Moses, Joshua etc. The selection of cdec over other MT systems is authors' | - Cdec's implementation of Hierarchical MT is used in this work. As we know, hierarchical decoding is also implemented in other MT systems such as Moses, Joshua etc. The selection of cdec over other MT systems is authors' | ||
- | - MIRA is used to train the MT system. | + | - MIRA is used for tuning feature weights. |
- Authors don't tune decoder in first-pass i.e. they don't calulcate feature weights (lambda) and probably they use weights from their previous experiments or setups. They don't clearly state the reason of this decision but our hypothesis is they might skiped the tuning step just to speed up the translation process. | - Authors don't tune decoder in first-pass i.e. they don't calulcate feature weights (lambda) and probably they use weights from their previous experiments or setups. They don't clearly state the reason of this decision but our hypothesis is they might skiped the tuning step just to speed up the translation process. | ||
- | - NIST-BLEU is used to compare results with official NIST evaluation whereas IBM-BLEU is used for evaluating the rest of experiments. We don't fully understand the use of different BLEU (prefering shorter sentences incase of NIST and longer incase of IBM) for evaluation and not sticking with NIST-BLEU | + | - NIST-BLEU |
- They gain maximum of 1.0 point increase in BLEU after combining all three features. | - They gain maximum of 1.0 point increase in BLEU after combining all three features. | ||
- Authors called BLEU as a " | - Authors called BLEU as a " | ||
- | i- They could have supported their argument by manually evaluating the test set. | + | |
- | ii- Instead of wasting half of the page length by criticizing over BLEU, they could have evaluated their system on other metric such as METEOR. | + | - Instead of wasting half of the page length by criticizing over BLEU, they could have evaluated their system on other metric such as METEOR. |
- | - Also, significance testing | + | - We believe that significance testing |
====Conclusion: | ====Conclusion: | ||
- | Paper is nicely written and all experiments are well documented. We believe that consistent translation choices system is better | + | Paper is nicely written and all experiments are well documented. We believe that consistent translation choices system is well suited only for translating from direction of morphologically-rich to morphologically-low language pairs and not the other way round. For translating |