This paper looks at how Synthesys provides real-time analysis of a fusion of structured and unstructured data. Synthesys brings together a diverse set of technologies into a seamless set of services with a rich API. Synthesys takes unstructured text as input, uses entity extraction with strong semantic relationship analysis to generate abstracted knowledge objects. These objects (people, places, connections, etc.) can then be used to understand and analyze what’s important to you in the data.
This paper describes the Natural Language Processing (NLP) capabilities in Synthesys, highlighting the ability to bootstrap high quality NLP and Named Entity Recognition (NER) models on an unfamiliar corpus (Lewis Carroll’s Alice in Wonderland). An overview of natural language processing (NLP), tooling, information on the model bootstrapping process, and discussions on the outcome and conclusions of the Alice in Wonderland experiment are presented.
In this paper, we explore the power of running Synthesys semantic analysis software on the Oracle Big Data Appliance to extract and resolve billions of entities, facts and relationships from millions of pieces of human language.