Last month, our engineering team hosted the latest instalment of the Yell hack day. Our hack days are fun ways for our large team of developers to break into smaller groups and brainstorm creative solutions around a particular theme. This session’s theme was data. Yell has one of the largest databases of businesses in the UK, add to that over half a million reviews and a large amount of visitor analytics and we had lots to be working with.
This hack day we saw seven teams of Yell developers enter with a wide variety of pretty interesting hacks and ideas to improve our products and the way we approach data at Yell. The day resulted in lots of interesting entries, from comparison tools to reading level analysis. The judge’s top three were:
Lee Dunkley, Steve Mitchell and Lance Ewing
Using gamification ideas, this team created an internal tool which had the potential to compare the content on Yell.com business profiles pages against their search engine relevance for a given query. This would show the user which features of their profile page give the most benefit to the business. While this comparison is interesting, the data could easily be misleading and we’re not certain on the science of it. It’s a nice experiment and we’re going to leave this one be for now.
Android Review Kiosk
Reviews are a great thing for both our customers and our merchants and our Android Developer Gavin came up with a photo booth-style Android tablet app that would allow you to make a quick review with your smile! Using your device’s camera and some face-recognition APIs, Gavin was able to update a user’s review rating based on how big a smile that person had – the bigger the smile, the higher the rating. This was a fun project that we’re not planning to roll out to the public, but the whole engineering team thought the idea was very, very cool.
Related Businesses Reviews
Mike Neeve, Natalia Sanchez Calderon, Simon Berry and Sam Lukes
Yell has a large database of reviews for its listed businesses and an active community of reviewers who are experts in their local area. This team wanted to know if users who reviewed bars also reviewed taxis and takeaways in the the same area. To do this they pulled out the entire set of reviews into a Neo4J graph database, connecting users with their reviews and locations. The team were then able to make connections between the various types of business, the reviews and the users.
This work is very exciting and is the most-promising piece of technology to come out of this hack day.
The Hack Day was a great time, we learned a lot from our team and we have some great ideas we may develop further in the future. Click through our gallery of images from the day, to see what we got up to.
Want to join our future Hack Days? Find out more about technology careers with Yell.