Go Back
Deep Learning

Processing

Humans have been recording and sharing written knowledge for thousands of years. Our brains are exceptional machines – built to learn, process, and communicate information in an unstructured way.

Natural Language Processing (NLP) is a subfield of artificial intelligence (AI) that focuses on the intersection of computers, organic language, and adaptive relationship-building. NLP strives to understand human language and is often associated with speech, image, and text recognition and interpretation.

Explored as early as the 1950s by the venerable Alan Turing in his article "Computing Machinery and Intelligence," NLP has captured the minds of researchers and practitioners for decades. The groundwork for NLP’s early stages appeared in handwritten "if-then" rule-based systems. Around the 1980s, these complex rule systems gave way to more exponentially-scalable statistical solutions. Statistical approaches now work from large data sets to produce dynamic language heuristics.

Thanks to a deep learning renaissance which began in the 2010s, NLP has taken a giant leap forward for language modeling and parsing.

Use Cases

Gravyty is powered by our patent-pending approach to NLP, which is weaved throughout our products.

Every single time a fundraiser interacts with Gravyty First Draft, we learn more about them as a professional. We track word choices, phrase patterns, context, and signatures to intuit how individuals speak and how organizations want to be portrayed. To synthesize more natural conversations, First Draft also introduces alternative words and concepts to continuously keep the language fresh.

Gravyty's algorithms learn how fundraisers engage with parents, alumni, board members, prospects, and more to tailor the appropriate conversational tone to the recipient. This adaptive model helps manage simultaneous email exchanges with diversely-motivated individuals. NLP allows Gravyty to build outbound communications that look and sound like our users much more quickly, regardless of the scenario or stage of the donor.

Resources

Ultimate Guide to Natural Language Generation

LINK

Improving Language Understanding by Generative Pre-Training

LINK

Building a Large-scale Commercial NLG System for an EMR

LINK

Recent Advances in Natural Language Generation

LINK