Abstract: 

Data scientists do not exist in a vacuum. To be useful, a data science team needs to communicate their results. This can happen through slide deques, through an API, through regular emails or through a dashboard. Interactive dashboards offer a particularly compelling way to present results: they allow the users to explore the data and models themselves, and discover their own trends.

Traditionally, creating dashboards required learning D3 and frontend technologies. Over the last few years, several packages have emerged for producing compelling dashboards directly in Python (Bokeh, Jupyter widgets, Plotly Dash) or R (RShiny). These lower the barrier to entry significantly for data scientists by allowing them to write dashboards in the language they are already familiar with.

In this talk, we give a brief overview of the options for writing dashboards in Python. We then show how to build a dashboard that allows users to upload a set of time-stamped tweets or sentences, performs sentiment analysis on the tweets, displays graphs of the evolution of sentiment over time and gives back a CSV of the sentiment attached to each tweet. We will write the dashboard live during the talk, using Plotly Dash.

We will then talk briefly about deployment options for data science dashboards, discussing infrastructure and security concerns.

By writing the dashboard live during the talk, we hope to give the audience a realistic flavour of what can be achieved and some insight into how to start. More advanced users will learn best-practices around writing dashboards.

Bio: 

Pascal Bugnion is a data engineer at ASI Data Science, a London-based data science consultancy. He works on building SherlockML, a collaborative platform for data science.

In his spare time, Pascal is an active contributor to open source. He helps maintain Jupyter widgets, a library for building user interfaces in Jupyter notebooks, and jupyter-gmaps, a tool for visualizing geographical data. He is the author of Scala for Data Science and holds a PhD in theoretical condensed matter physics from Cambridge University.