07.05.2019

PRISME: structured data sharing facilitated

by Bram Weytjens

This post emphasises how XAOP facilitates data sharing efforts in an industrial setting by providing an application which focusses on gathering data while retaining data quality for hierarchically structured data

About Bram

Bram is all about connections. With a doctorate in bio-informatics on network-based analysis, he loves turning data into graphs so even people without doctorates can easily understand them. When he’s not in front of his computer, you can find him in his basement brewing beer. We should get a hold of some of those beers soon to find out if they’re any good.

Idea

The PRISME FORUM organisation (https://www.prismeforum.org/) is a collaborative group consisting of R&D IT leaders from more than 25 biopharmaceutical companies which can involve as many as 100 people in its activities. The goal is to share non-commercially sensitive information with one another in order to enhance the efficiency, effectiveness and impact of global research and development information management and information technology organization within the biotechnology and pharmaceutical industries.

XAOP provides a survey application to the PRISME FORUM organisation. As the number of members and the amount of data increased, the need for an application to manage these data streams with a focus on data quality and data visualisation arose.

As the data is hierarchically structured in a “business ontology”, which reflects to what domain, subdomain, sub-sub… of business a specific piece of data belongs, a classical survey application did not suffice. Furthermore, people tend to make spelling mistakes when filling out surveys, leading to difficulties and -often large- data cleanup efforts during data visualisation.

Approach

To accommodate these needs, XAOP created a survey application in which the business ontology is a central aspect. Admins are able to alter the business ontology used for a specific survey inside of the applications. Users on the other hand can fill in and filter their data based on the ontology, which allows them to easily navigate through their data as specific users are often knowledgeable about data related to specific parts of the business ontology. By looking at past data, we set up fuzzy search algorithms which suggest the correct spelling whenever a user inputs data. This practice increases data quality and reduces data cleaning efforts significantly.

As the resulting data is well-structured and clean, intuitive data visualisations are easy to achieve and the time to insights is reduced.