Difference between revisions of "Lingala and Kikuyu/Structural Transfer"

From LING073
Jump to: navigation, search
Line 2: Line 2:
  
 
==Pre-Evaluation==
 
==Pre-Evaluation==
 +
Below are the evaluation statistics for both the mini test corpus and the full parallel corpus:
 +
===lin → kik evaluation===
 +
Evaluation of mini corpus:
 +
* WER: 87.50%
 +
* PER: 87.50%
 +
* Percentage unknown words: 0.0%
 +
* Percentage known words: 100%
 +
 +
Evaluation of full corpus:
 +
* WER: 99.06%
 +
* PER: 94.34% (unknown-word marks removed)
 +
* Percentage unknown words: 66.67%
 +
* Percentage known words: 33.33%
 +
 +
===kik → lin evaluation===
 +
Evaluation of mini corpus:
 +
* WER: 80.00%
 +
* PER: 80.00%
 +
* Percentage of unknown words: 0.0%
 +
* Percents of known words: 100%
 +
 +
Evaluation of full corpus:
 +
* WER: 98.13%
 +
* PER: 90.65% (unknown-word marks remove)
 +
* Percentage of unknown words: 70.75%
 +
* Percentage of known words: 29.25%
  
 
==Implementation==
 
==Implementation==
  
 
==Post-Evalatuion==
 
==Post-Evalatuion==
 +
===lin → kik evaluation===
 +
Evaluation of mini corpus:
 +
* WER: ?
 +
* PER: ?
 +
* Percentage unknown words: ?
 +
* Percentage known words: ?
 +
 +
Evaluation of full corpus:
 +
* WER: ?
 +
* PER: ? (unknown-word marks removed)
 +
* Percentage unknown words: ?
 +
* Percentage known words: ?
 +
 +
===kik → lin evaluation===
 +
Evaluation of mini corpus:
 +
* WER: ?
 +
* PER: ?
 +
* Percentage of unknown words: ?
 +
* Percents of known words: ?
 +
 +
Evaluation of full corpus:
 +
* WER: ?
 +
* PER: ? (unknown-word marks remove)
 +
* Percentage of unknown words: ?
 +
* Percentage of known words: ?
  
 
[[Category:Sp17_StructuralTransfer]]
 
[[Category:Sp17_StructuralTransfer]]
 
[[Category:Lingala]]
 
[[Category:Lingala]]
 
[[Category:Kikuyu]]
 
[[Category:Kikuyu]]

Revision as of 17:13, 2 April 2017

This page details transfer rules for translation between Lingala and Kikuyu

Pre-Evaluation

Below are the evaluation statistics for both the mini test corpus and the full parallel corpus:

lin → kik evaluation

Evaluation of mini corpus:

  • WER: 87.50%
  • PER: 87.50%
  • Percentage unknown words: 0.0%
  • Percentage known words: 100%

Evaluation of full corpus:

  • WER: 99.06%
  • PER: 94.34% (unknown-word marks removed)
  • Percentage unknown words: 66.67%
  • Percentage known words: 33.33%

kik → lin evaluation

Evaluation of mini corpus:

  • WER: 80.00%
  • PER: 80.00%
  • Percentage of unknown words: 0.0%
  • Percents of known words: 100%

Evaluation of full corpus:

  • WER: 98.13%
  • PER: 90.65% (unknown-word marks remove)
  • Percentage of unknown words: 70.75%
  • Percentage of known words: 29.25%

Implementation

Post-Evalatuion

lin → kik evaluation

Evaluation of mini corpus:

  • WER: ?
  • PER: ?
  • Percentage unknown words: ?
  • Percentage known words: ?

Evaluation of full corpus:

  • WER: ?
  • PER: ? (unknown-word marks removed)
  • Percentage unknown words: ?
  • Percentage known words: ?

kik → lin evaluation

Evaluation of mini corpus:

  • WER: ?
  • PER: ?
  • Percentage of unknown words: ?
  • Percents of known words: ?

Evaluation of full corpus:

  • WER: ?
  • PER: ? (unknown-word marks remove)
  • Percentage of unknown words: ?
  • Percentage of known words: ?