H2O.ai Expands Driverless AI to New Class of Use Cases with Natural Language Processing
Ideal for Applying AI to Text-Centric Applications
H2O.ai, the open source leader in AI, today announced a new release of its award-winning automatic machine learning platform H2O Driverless AI, which includes advanced Natural Language Processing (NLP) capabilities. The latest innovations were spearheaded by the #1 Kaggle Kernels Grandmaster, Sudalai Rajkumar “SRK”, in Chennai, India, alongside H2O.ai teams around the world. These additional NLP capabilities and integrations will enable organizations to expand their current AI strategies, directly address key machine learning use cases, improve the accuracy of many predictive models like fraud detection and churn, and expands the use to sentiment analysis, document classification and other text-centric applications.
With the addition of NLP recipes, Driverless AI can now handle even more types of data right out of the box, saving organizations time and money by making it easy to create machine learning models that using data that contains blocks of text, numeric and categorical data. In the latest version, larger volumes of text data, such as description fields, are used directly by the platform, saving data scientists time traditionally required to convert that text into predictive features. H2O.ai has further integrated NLP with TensorFlow which provides a deep learning approach which is helpful for a variety of problems and enhances Driverless AI’s NLP capabilities to process larger volumes of text automatically.
“Text is a uniquely human expression – making natural language processing the ultimate AI challenge. Our new NLP for text recipes in Driverless AI using TensorFlow and machine learning expands the class of problems and use cases that autoML can solve for enterprises in several domains,” said Sri Ambati, CEO and founder at H2O.ai. “Text is naturally intertwined in enterprise data, and data scientists are being increasingly expected to train learning architectures for semi-supervised and unsupervised challenges. Driverless AI provides faster, cheaper and easier way to train and reuse deep learning text models. It’s like having an NLP expert on your team.”
With new NLP recipes, H2O Driverless AI customers can now:
- Better predict buying patterns for retail based on customer reviews
- Do higher quality equity research on reports for securities
- Classify documents based on their text content alone
- Determine customer sentiment from the transcript of a customer service call to determine the likelihood of customer churn
- Enhance existing models for fraud, pricing, marketing and more by including textual data like product descriptions and customer reviews
- Save lives with domain-specific NLP for healthcare
H2O Driverless AI empowers data scientists or data analysts to work on projects faster and more efficiently by using automation and state-of-the-art computing power to accomplish tasks that can take humans months in just minutes or hours by delivering automatic feature engineering, model validation, model tuning, model selection and deployment, machine learning interpretability, time-series, NLP and automatic pipeline generation for model scoring.