Last week our CEO, Filip Stachura, gave a speech at the EARL conference in Boston titled “Rapid shiny development to blend experts knowledge into machine learning models”.
Filip spoke to a huge number of R data scientists about how one can improve algorithmic models with a human-experts knowledge. Such approach still delivers better results than pure machine learning models or manual work.
Filip presented how we at Appsilon Data Science work on R based project. First of all… we love R and Shiny. We amaze our customers by delivering working dashboards in the first days of our cooperation. Thanks to that our clients are able to apply newly discovered insights to improve decision-making process almost immediately. What’s more, we provide them with a User Interface, which is much more pleasant and intuitive to work with than Excel sheets.
At EARL Filip also presented our two internal extensions for Shiny for building even more beautiful and reactive apps. The first one allows us to use SemanticUI web library components while the second one is a wrapper for No-SQL real-time databases, like RethinkDB or MongoDB. This allows us to build a distributed R app with realtime collaboration. We plan to release them as open-source libraries in the first quarter of 2017.
Filip made his presentation in R Markdown, where he embedded a sample app utilising both libraries and we want to share it with you all. Check it out:
Feel free to get in touch, if you want to know more about this topics.
NOTE! in this presentation you can navigate using keyboard arrows ← → or swipe gesture)