Abstract: According to Tom Standage's innovative account of world history, coffee is one of the 6 beverages that has shaped the world. Indeed from fuelling the rise of the Dutch East Indies company to powering multi-million dollar businesses in the 21st century, coffee has shaped geographies and businesses. Geospatial patterns spawned as a consequence of geopolitics, socio-economic dynamics and even climate change impacts at different scales percolate down to your cup of coffee. I present a business use case-"Selecting the Most Optimal Location For a New Cafe in San Francisco" in which we will see how we can deploy data science-based techniques such as web scraping to extract information about the geo-locations of San Francisco's existing neighbourhoods and their cafes. Using publicly available APIs provided by FourSquare and/or Twitter we will map the most popular existing cafes in these areas and obtain their demographic and/or socio-economic variables using public databases. After establishing where the most popular cafes are, their broader geospatial and socioeconomic context and what makes them tick, we will scope neighbourhoods where a new cafe could be opened. Finally, we will look at the broader picture and quantify how climatic factors and geopolitics could influence the coffee supply chains (and subsequently our cafe). With this, we will also look at if and how we can lean into resilient and sustainable supply chains to make our cup of morning coffee resilient to external shocks and make sustainable coffee a winning business strategy.
Bio: I completed a PhD (University of Cambridge, UK) in 2017 where I focussed on implementing data science techniques for quantifying the impact of forest loss on tropical ecosystems. I hold an MPhil (School of Geography and Environment) and an MSc (Department of Engineering) from Oxford University. I have more than 10 year's experience in conducting academic research (published in high level peer-reviewed international scientific journals such as PLOS One) and advising both non-governmental and industry stakeholders in data science, deep learning and earth observation (EO) related topics.
I have a strong track record in implementing machine learning, data visualization, spatial data analysis, deep learning and natural language processing tasks using both R and Python. In addition to being educated at the best universities in the world, I have honed my statistical and data analysis skills through many MOOCs, including The Analytics Edge (R based statistics and machine learning course offered by EdX), Statistical Learning (R-based Machine Learning course offered by Stanford online) and the IBM Data Science Professional certificate Track. I specialise in a variety of topics ranging from deep learning (Tensorflow, Keras) to machine learning to spatial data analysis (including EO data processing), data visualizations, natural language processing, financial analysis among others. I have acted as a peer reviewer on highly regarded academic journals such as Remote Sensing.