Computer systems for classifying multilingual text

A method for identifying textual information regarding a first topic, comprising:

• accessing a plurality of text units;
• detecting that a first text unit of the plurality of text units is in a first language;
• accessing a first language keyword set for the first topic, the first language keyword set comprising a first plurality of keywords associated with the first language;
• determining a first relevance score for the first text unit based at least in part on the first language keyword set;
• determining that the first relevance score is greater than a relevance score threshold;
• and
• translating the first text unit to a base language;
• and
• determining a text unit topic for the first text unit using a classification model trained with training data in the base language.

A machine-readable medium comprising instructions thereon that, when executed by at least one processing unit, cause the at least one processing unit to perform operations comprising:

• accessing a plurality of text units;
• detecting that a first text unit of the plurality of text units is in a first language;
• accessing a first language keyword set for a first topic, the first language keyword set comprising a first plurality of keywords associated with the first language;
• determining a first relevance score for the first text unit based at least in part on the first language keyword set;
• determining that the first relevance score is greater than a relevance score threshold;
• and
• translating the first text unit to a base language;
• and
• determining a text unit classification for the first text unit using a classification model trained with training data in the base language.