Logistics & Transportation


Goal. A global e-commerce company wanted to improve prediction accuracy for their shipments within the EU. Analytics provided visibility into the largest gaps between predicted and actual shipments.

Methods. I began by joining previously siloed data from each region and gathered stakeholder requirements, such as how key metrics are defined. A combination of Python and MySQL ensured much faster processing than their status quo.

Results. A range of performance charts provided flexible, EU-wide visibility to uncover links that were previously hidden in the data. This allowed the company to improve daily predictions and boost operational efficiency.


High Tech

San Francisco, CA

Goal. This high-tech start-up wanted to map trends in the global food space in order to inform their client’s strategic decisions. Based on publicly available news articles, data analytics revealed semantic similarities in thousands of articles to draw a coherent map of topics.

Methods. In close collaboration with the client, the data scope was defined to include relevant geographies. Entities such as people, organizations and topics were extracted from each news articles to compute the similarities. This enabled a powerful visualisation of how topics like “ready-meals”, “low-calorie diets” and “organic food” have evolved and are related to one another.

Results. Decisions that would have taken a couple of months and a large budget were now made much quicker and with higher confidence due to analytics.

Sounds interesting?

Find out how Machine Learning can help your organisation in a free 10-minute Skype call.