Issue #30 – Reducing loss of meaning in Neural MT

28 Mar19

Author: Raj Patel, Machine Translation Scientist @ Iconic

An important, and perhaps obvious feature of high-quality machine translation systems is that they preserve the meaning of the source in the translation. That is to say, if we have two different source sentences with slightly different meanings, we should have slightly different translations. However, this nuance can be a challenge, even for state-of-the-art systems, particularly in cases where source and target languages partition the “meaning space” in different ways. In this post, we will try to understand what are the language characteristics causing this meaning loss and discuss a few methods on how to reduce it in Neural MT.

Many-to-One Translation

Translation systems are typically many-to-one functions from source to target language, which in many cases results in important distinctions lost in translation. For example, take the following sentences:

1. By mistake, I cut off my finger with a knife.
2. By mistake, I cut my finger with a knife.

Using Google Translate for English to Hindi, they both are translated in the same way (as below) which is ambiguous in Hindi:

• “गलती से मैंने चाकू से अपनी उंगली काट ली। (galatee
  To finish reading, please visit source site