Best of 2019: Our top 10 most read content
If you were following along with us online this past year, you may have learned how to solve a Rubik’s cube. We also demonstrated how to utilise network science for a new approach to recommender engines.
But, did you also know we saved employees with the municipality of Utrecht countless hours by optimising their PDF search? Or, that we’ve helped HEMA figure out who’s buying rookworst in the rain?
In case you missed any of these reads, you’re in luck. These stories all topped the list of our most popular articles and case studies in 2019. Here’s the full round up of what our readers found most worthy to read.
Top 5 articles
Mats Valk is a data scientist with an unusual passion – solving a Rubik’s cube as quickly as possible. He answers the age old question for us: How on earth do you solve a Rubik’s cube?
Municipalities are faced with a myriad of issues. Housing, education, safety, and the environment are all integral to the well being of citizens. But, cities aren’t the only stakeholders and such complex social issues extend well beyond the boundaries of any single organisation. With the Datalab method, municipalities and social partners come up with better solutions for complex social issues based on data.
Struggling with uncertainty? Why not start simple. Data Scientist Yu Ri explains how you can calculate different types of uncertainty for regression problems using quantile regression and Monte Carlo dropout.
ProRail reached out to Xomnia to help them more accurately predict how long it takes to resolve an incident causing a train delay. They wanted to provide better information on the most logical steps to take leading up to the resolution. We accepted the challenge, but soon discovered a major flaw that derailed the entire approach.
Recommendations that actually match customers’ needs or desires can increase conversion rates and unlock extra revenue at checkout. That’s why well designed recommender engines are very desirable for webshops. Junior data scientist Max demonstrates his creative approach to recommender engines.
Top 5 case studies
The challenge of providing customers with the correct contact information for customer service questions when you have 21,000 selling partners was time-consuming and manual at Bol.com. Automating this process helped the company provide a faster and more intuitive experience for its customers while cutting costs for their customer service operations.
What if an airline could predict delays in advance? They’d be able to cut their losses and address the issue proactively. That means fewer delays and happier customers. Here’s how we made it happen for KLM.
For the Municipality of Utrecht, searching through a digital archive containing 10,000 scans of PDF documents from over 500 real estate properties including drawings was a technical challenge, and quite a headache. That is, until we designed and implemented for them a smart PDF search engine comparable to Google.
NS International wants to help people be more environmentally friendly by travelling Europe by train. Our junior data scientists helped NS International with quality analysis of their customers’ website experience.
Category and store managers of HEMA were still drawing on topline KPI’s, their own intuition and weather to make predictions or explanations. This is how we helped revolutionise the company’s approach to sales forecasting and budgeting, stock management, in-store and online promotions, and performance management.