Epiq uses Reveal’s language identification technology on a multi-language document review matter – enabling them to segregate documents for separate English and Japanese review workstreams.
Challenges
- Documents contain multiple languages and sometimes a combination of both. For example, an email in Japanese may have English attachments
- Case team desired to use separate review teams to focus on each language
Solution
Reveal’s language identification technology was used to quickly segregate documents based on language, enabling the Client to properly match documents with reviewers by language. About 25% of the document population was sent to the Japanese team – including documents where email families contained a mix of languages.
The Client’s team then augmented the review process using Reveal’s active learning technology to identify highly relevant documents for outside counsel.
Results
Using AI engines to separate a document collection based on predominant language was an effective method for using language-specific reviewers. The client was able to start the review process sooner and more accurately send relevant data to the proper teams.
AI-powered review eliminated the need to review more than 300,000 documents and provided a direct savings to the LSP’s client of approximately $560,000.