Case Study: Hootsuite achieves fast, accurate social media language detection with Basis Technology's Rosette

Rosette Delivers Accurate and Fast Language Detection for Social Media Monitoring

Hootsuite, the social media management platform used by more than 18 million people and 80% of the Fortune 1000, needed a better way to segment social media messages by language for regional and customer support analysis. Its home-grown language detector required frequent refreshes to keep up with evolving vocabulary, creating maintenance overhead and risking misdirected posts and support requests. Basis Technology’s Rosette Language Identifier was brought in to improve language detection across Hootsuite Insights and Analytics.

Basis Technology’s Rosette Language Identifier uses statistical language profiles and n-gram analysis to accurately identify 55 languages, including short social posts like tweets and captions. The solution delivered fast, precise language tagging so Hootsuite could route messages to the right teams, assess staffing needs by language, and reduce time spent maintaining its own detector. Hootsuite reported that the precision is strong enough that they rarely revisit the decision to use Rosette.

Hootsuite

Mihai Caraman

Development Manager