[ Skip to the content ]

Institute of Formal and Applied Linguistics Wiki

[ Back to the navigation ]

You are here: start » courses » rg » transductive_learning_for_statistical_machine_translation

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision 		Previous revision
Next revision Both sides next revision
courses:rg:transductive_learning_for_statistical_machine_translation [2010/12/08 22:00]
jawaid 		courses:rg:transductive_learning_for_statistical_machine_translation [2010/12/08 22:27]
jawaid
Line 31: Line 31:
   * Data filtering is performed on both bilingual and monolingual data to keep only that part of the data which is relevant to the test data.   * Data filtering is performed on both bilingual and monolingual data to keep only that part of the data which is relevant to the test data.
  
-  * They used three different evaluation metrics for evaluating translation qulaity: **BLEU** score, **mWER** (multi-reference word error rate) and **mPER** (multi-reference position-independent word error rate).+  * They used three different evaluation metrics for evaluating translation quality that are: **BLEU** score, **mWER** (multi-reference word error rate) and **mPER** (multi-reference position-independent word error rate). 
 + 
 +  * Experiments are performed on EuroParl and NIST corpus. On EuroParl corpus the selection and scoring was carried out using importance sampling with normalized score. Three different experiments are performed on the Europarl corpus which didn't produce significant improvement in the accuracy on output translation. Experiments on NIST data are performed using three different test sets where the output translation of each test set yield the best score when threshold-based selection method was combined with confidence estimation as scoring method. 
 + 
 +  * The main issue with this paper is that number of iterations that are selected to train the model are not described. And according the Figure 1 in the paper graph achieves global maximum on iteration 16 and iteration 18 on train100k and train150K corpus. Where our main concern is they might know the BLEU score from the test set they already have and they stopped the training process or cut down the graph after particular iteration where BLEU is optimized according to already computed BLEU score.
  
   *    * 
Line 49: Line 53:
 ===== What do we dislike about the paper ===== ===== What do we dislike about the paper =====
  
-  * The technique defined in the paper uses the source side test data during training process which limits the use of this technique in different applications. For instance this learning mechanism can not be applied in online translation systems.+  * The semi supervised learning scheme presented in this paper uses the source side test data during training process which limits the use of this technique in different applications. For instance this learning mechanism can not be applied in online translation systems. 
 + 
 +  * 
  
  
    
  

[ Back to the navigation ] [ Back to the content ]