Why we translate

At TalentNeuron, much of the labor market data we provide is derived in whole or in part from job postings. We gather job postings from thousands of sources across 32 countries (and counting).

Job postings are usually written using the language of the region in which the position is located. In order to analyze these job postings (to identify the required skills, for example), we must first translate those local languages into English (the language of our products). 

Currently, we translate from the following languages:

  • Arabic

  • Chinese (Mandarin)

  • Czech

  • Danish

  • Dutch

  • Finnish

  • French

  • German

  • Hungarian

  • Italian

  • Japanese

  • Korean

  • Malay (Bahasa Melayu)

  • Norwegian

  • Polish

  • Portuguese

  • Romania

  • Spanish

  • Swedish

  • Thai

  • Turkish

How translation works

We translate using a neural net algorithm. In contrast with rules-based translation models (which compare texts against long lists of manually-established rules), neural net translation uses machine learning to learn correct translations through real-world examples. That is to say, the model is trained on a large set of sample texts, which allows the translation to be more sensitive to context and allows for more natural and logical translations.