Improving Organizational Knowledge with Natural Language Processing Enriched Data Pipelines

Improving Organizational Knowledge with Natural Language Processing Enriched Data Pipelines

Thursday, May 23
2:00 PM - 2:40 PM
Marquis Salon 9

The information age has allowed everyone to tap into the exponential production of data. Unfortunately, much actionable insight is the result of unexpected or anomalous behavior that can only be recognized through experience. A collection of NLP microservices was crafted to complement an organization’s existing technology infrastructure in order to translate and bring additional meaning to an organization’s already existing and real time collection of unstructured text.

In this session, and in collaboration with Partners & Co., a Chicago-based real estate firm, we will demonstrate how we can leverage an organization’s collective knowledge and turn unstructured text that is generated from across various communication mediums into real time actionable insight. We will demonstrate how we can use a combination of open source tools such as Apache NiFi, Kafka, OpenNLP, and Superset to build a full streaming NLP pipeline to consume unstructured text, detect the language and sentences within the text, deconstruct the grammatical makeup, and derive meaning of the entities identified within the text.

SPEAKERS

Jeff Zemerick
Cloud and NLP Consultant
Mountain Fog, Inc.
Jeff is a software engineer and cloud architect. He is a committer and PMC on Apache OpenNLP. Jeff currently works on natural language processing pipeline projects and resides outside of Morgantown, WV.
Eric Wolok
Managing Director
Partners & Co.
At Partners & Co., Eric Wolok specializes in the sale of commercial real estate in Chicago. Partners & Co. uses open source tools such as emacs, sed, awk, Apache NiFi and Apache Spark to identify, track and facilitate unique investment opportunities for their clients.