Developing a Warning System for Risk Managers from Scratch on GCP, using AI & ML (Cloud Next '18)

Опубликовано 26 июля 2018, 19:41

In this session we will demonstrate a Proof of Concept Early Warning System for Credit Risk Managers, built on the Google Cloud Platform. The flexibility of the GCP enabled us to develop this PoC from scratch in a timeframe of 12 weeks. The demo will feature how counter-parties (for a financial institution) can be selected, how various data sources (market, news, etc.) are ingressed (GDELT, relevant data vendors), processed using Pub/Sub and Container Engine. Further demo how data and news is analyzed/translated/filtered and clustered involving a range of APIs a.o. Natural Language API utilizing features like entity recognition, sentiment analysis, and content classification. Results displayed based on relevance score tuned to needs of Credit Risk Managers enabling them to focus on only the most relevant information.

MLAI204

Event schedule → g.co/next18

Watch more Machine Learning & AI sessions here → bit.ly/2zGKfcg
Next ‘18 All Sessions playlist → bit.ly/Allsessions

Subscribe to the Google Cloud channel! → bit.ly/NextSub


event: Google Cloud Next 2018; re_ty: Publish; product: Cloud - General; fullname: Otto Vermuelen;