RWS: saving time and money with incident report processing

Dealing with massive amounts of report data, and extracting relevant information from them, is time-consuming and costly. In this project we smartly automated these tasks with entity extraction and document classification.

Challenge

Locating and classifying relevant mentions of an entity in a piece of text – called entity extraction – is a smart way to do away with the manual processing of report data. The job in itself is repetitive, straining, and costly as well.

Solution

We automate the extraction of entities from incident reports and plot them, by creating a machine learning pipeline that auto-labels the incidents. A smart UX interfaces between AI and team members, and helps speed up the processing. Added benefit: the team gets more time to focus on key tasks.

Result

In production. New version with additional features on a yearly basis.

.css-l0mio9{display:none;visibility:hidden;}