How Neural Machine Translation Will Change E-Discovery

When a global enterprise gets sued, it’s vital to know who is involved and how. But finding out who to blame isn’t always simple.

Global law firms are tasked with sifting through thousands, sometimes millions of emails, chats, and legal documentation during eDiscovery. These documents and audio recordings could be in many different languages and stored around the world. Sometimes that data is stored in countries with strong data protection regulations, such as Brazil and parts of the EU, so it cannot under any circumstances leave the country.

So, how can an office in the U.S. review hundreds of days of correspondence in multiple languages?

If the firm hires translators, they’ll need dozens with a strong knowledge of everything from slang to deep subject matter expertise of the topic in discovery. If instead they decide to go with an e-discovery translation solution, they’ll still need help during the review process, especially for data in Asian languages – there are several ways to interpret one word, for which there may be five slang alternatives. In either case, the team must spend a lot of time and money to get reliable and accurate results.

Until now, that is.

PNMT is the tipping point in a technology that’s been optimized for ~50 years

Neural machine translation (NMT) runs on neural networks, the architecture that all artificial intelligence is made of. The status quo in translation has been that you must sacrifice time and money for accuracy, or accuracy for time and money. With NMT, however, you can get fast, accurate translations that are both affordable and quick.

Here’s how: Based on our own brain’s neural network, NMT can do what other machine translation methods have not done before – it achieves translation of entire documents without losing meaning. As you know, a misinterpreted word or sentence can have devastating consequences to a lawsuit. Neural machine translation software, such as Pure Neural Machine Translation (PNMT) from SYSTRAN Software, views sentences as a whole instead of translating a few words at a time (this is why the traditional statistical machine translation was called ‘phrase-based’ translation). This new approach allows the software to translate sentences in the context of an entire paragraph or document for a more accurate translation.

PNMT has shown large gains in Asian languages where it’s important to see the whole sentence to understand the context.

In a recent testimonial from a PNMT beta tester who needed to translate a 59-paged technical presentation quickly, Lori Thicke, Found & CEO at Lexcelera, remarked, “Usually with machine translation, there is a lot to correct unless you have trained the engine very, very well… But this time, the SYSTRAN PNMT engine was a completely untrained, generic engine. And still, the translation of our company presentation was amazing.”

How it works in real-life scenarios

Imagine you’re asked to go through 320GB of content in Chinese for a global manufacturer. To do this, you need to translate the documents into the target language to find relevant information, and you need to do it quickly. Your firm would typically find a lawyer with engineering expertise, who is fluent in Chinese and English. This subject matter expert is very expensive and charges hourly. Alternatively, using human translators to transpose data will take you weeks and hundreds of thousands of dollars.

With PNMT you can automatically translate terabytes of documents. Translation software running on a NMT engine can stand alone or be integrated with your eDiscovery platform, such as CMLess for Relativity, to assist in reducing your review costs and improve the efficiency and effectiveness of technical assisted review (TAR). These tools replace up to 50% of the work that human reviewers or native translators do. With this in mind, they can focus on finding, translating and analyzing important documents based on a set of keywords and phrases.

SYSTRAN translation solutions can also improve security by limiting the number of people viewing sensitive documents. Further, as a peripheral benefit during your day to day global collaboration, our enterprise translation includes plugins that work within Outlook, Word, Excel or PowerPoint. It can also be installed behind your firewall, within your intranet, keeping translations of sensitive data offline.

The edge that a law firm gains from being able to discover evidence and answer questions securely in real-time provides a significant ROI in time and money for both lawyer and client. This is why SYSTRAN works with over 60% of the providers listed in Gartner’s eDiscovery Magic Quadrant.

See for yourself

In a more connected and ever transforming world of risk and opportunity, SYSTRAN NMT provides a new standard for communication across countries and cultures. Utilizing PNMT software for eDiscovery translation can give you the ability to move faster, process high volumes of documents quicker and find what you need with greater efficiency than ever before.

Discover more about SYSTRAN’s PNMT engine for foreign language eDiscovery.