Cross-lingual search based on concepts and meaning

Cross-lingual search would be a snap if the search engine could read your mind. It would then perfectly translate your intent and find exactly what you are looking for in multiple languages. While we are not yet at search utopia, there is now semantic search that uses word meaning instead of “just words” to search across languages.

Until recently, machine translation was the primary method of searching across languages either by translating search keywords into other languages or translating searchable records into English.

However, machine translation, or for that matter human translation, loses the valuable nuances and meaning of the original text.

This white paper explores an approach that delivers better accuracy based on semantics (meaning) to retrieve ideas. First, we will compare the traditional translation-based approach with semantic similarity. Then we will see what it takes to implement semantic search in a production environment, looking at use cases in topic and event extraction and cross-lingual name matching.

Download the whitepaper today to learn more.