Extracting Essential Meaning in Multi-Language e-Discovery

October 12, 2022 • 

Until now, keyword search has been the main method for finding relevant documents from mountains of data in e-discovery. But keyword search requires huge lists of terms, which has the effect of limiting its scope and precision. It catches too little, e.g., by focusing on a specific spelling while missing key alternatives, like different word forms and other ways of saying the same thing. More critical, keyword search doesn’t understand context or meaning.

What if your documents are in multiple languages? You’d need those keyword lists — translated — for all languages. And you’d have to be confident the keyword terms were correctly translated and interpreted for each language. There has to be a better way to search documents in multiple languages and find the ones that contain precise, focused meanings relevant to your case.

Now there is. Through a partnership between BasisTech and Ai Translate by LSI, you can use a powerful, AI-driven methodology that will find and extract key documents containing your search phrases or concepts across multiple languages. This process uses cross-lingual semantics to find the precise meaning-based content relevant to your case. It keeps your key concepts from being lost in machine translation while boosting the speed and accuracy of e-discovery. So, rather than translating reams of documents first, you can focus translation efforts on the files that are actually pertinent to your case.

In this webinar, we describe how to take text analytics to the next level for e-discovery. You’ll learn how you can search across languages — without needing in-house competency in those languages — to improve accuracy, bolster your case, and potentially slash your e-discovery time from months to weeks, or even days. You’re taking control of your matter earlier before more costly bilingual staffing decisions are made.

See also the companion blog post “How text analytics and machine translation made multilingual eDiscovery 151% more productive”

Extracting Essential Meaning in Multi-Language e-Discovery


Eugene S. Reyes

Federal Solutions Engineer

Eugene Reyes is a Federal Solutions Engineer at BasisTech. He has decades of experience in software engineering, mainly in several offices within the United States Intelligence Community (IC) and Department of Defense (DOD). Mr. Reyes has been part of multiple applied research and development organizations, focusing on advancing text analytics through various Human Language Technologies (HLT), such as Latent Semantic Indexing (LSI), entity extraction, and machine translation. At BasisTech, he created Semantic Studio to not only demonstrate existing BasisTech product capabilities, but also to showcase the “art of the possible” with semantics and more.

Prior to joining BasisTech, Mr. Reyes was a Software Engineer with Numbers Place, Inc., where he architected and developed an HLT solution for the IC that exceeded client needs, discovered previously-unknown topics and identified critical files in a timely manner. This solution earned him numerous IC awards for innovation and collaboration, and his efforts were honored with meritorious unit citations. Before that, Mr. Reyes was a software engineer for Agilex Technologies and Raytheon, where he won multiple awards for leadership and dedication. Growing up, he learned to program at an early age, lectured programming classes, tutored students, and led multimedia productions. Mr. Reyes obtained a B.S. in computer science and a minor in political science at Virginia Polytechnic Institute and State University.

Jason E. Boro, Esq.

Strategic Partnership Development Manager
Linguistic Systems, Inc.

As Strategic Partnership Development Manager at Linguistic Systems, Inc., headquartered in Boston, MA, Mr. Boro manages research and development of new products. Mr. Boro possesses over 15 years of eDiscovery experience. Prior to LSI, Mr. Boro spent 4 years with the U.S. Attorney's Office in Massachusetts. He has also practiced law in Massachusetts; in the Federal District Court for Massachusetts; and in Washington, DC. Mr. Boro holds a graduate degree in Linguistics, from the University of Connecticut. He also completed a course of study and conducted research elaborating psychological models for parsing sentence grammar while supplying a theoretical account of the maturation of innate syntax.