Monthly Archives: April 2017

Word-sense disambiguation: first test of new engine

Now testing the new engine with the semantically ambiguous French ‘échecs’ = fiaschi/scacchi (failures/chess).

What is interesting here is that semantic disambiguation transfers successfully into English (although the French/English engine is still in its infancy as there are still a lot of grammatical errors):

Now further tests are needed with some other semantically ambiguous words:

  • ‘défense’: defense/tusk; Corsican: difesa/sanna
  • ‘fils’: sons/wires; Corsican: figlioli/fili
  • ‘comprendre’:
    understand/comprise; Corsican: capisce/cumprende
  • ‘vol’: flight/theft; Corsican: bulu/arrubecciu
  • ‘voler’: fly/steal; Corsican: bulà/arrubà
  • ‘échecs’: chess/failures; Corsican: scacchi/fiaschi
  • ‘palais’: palace/palaces/palate/palates; Corsican: palazzu/palazzi/palate/palates

In the background, the unresolved threefold ambiguity of French ‘partie’ = parti/partita/partita (part/game/gone) is lurking…

Feigenbaum test and semantic disambiguation

Now it is patent that there cannot be successful  Feigenbaum test (i.e. not only occasional Feigenbaum hits, but regular and average performance) without an adequate treatment of semantic disambiguation. Arguably, it is one hard problem of machine translation. Here are some typical instances:

  • ‘défense’: defense/tusk; Corsican: difesa/sanna
  • ‘fils’: sons/wires; Corsican: figlioli/fili
  • ‘comprendre’:
    understand/comprise; Corsican: capisce/cumprende
  • ‘vol’: flight/theft; Corsican: bulu/arrubecciu
  • ‘voler’: fly/steal; Corsican: bulà/arrubà
  • ‘échecs’: chess/failures; Corsican: scacchi/fiaschi
  • and the fourfold ambiguous ‘palais’: palace/palaces/palate/palates; Corsican: palazzu/palazzi/palate/palates

In short: no successful semantic disambiguation = no genuine successful  Feigenbaum test. Semantic disambiguation engine needs to be rewritten.

French to English: superlative

Now testing French to English translation. Still a lot of grammatical errors. The scoring is a rough 80%. Adjective-noun order is now handled properly. But some progress in superlative is expected:

  • ‘le plus important’ should translate: the most important
  • ‘le plus gros’ should translate: the largest

To notice: the disambiguation of French ‘plus’ did worked properly since ‘plus’ is ambiguous between ‘more/most’ and ‘pleased’.

Adjective-noun reversal in English

Now beginning to incrementally improve the translation from French to English. Beginning with adjective order in noun + adjective structures:

  • ‘présence significative’ = significant presence
  • ‘soldats coloniaux’ = colonial soldiers
  • ‘colonies françaises’ = French colonies