Imagine you want to open a new store, and need to pick the right location for it. In order to minimize risk, you can select the right conditions analyzing multivariate geolocated data. Once there, you can train ML models to suggest to you the perfect spots. Let’s check how to do all that in Python!Abstract
80% of the data generated and collected today contains location information. Most of our activities using GPS systems, mobile apps and IoT devices produce geospatial data that organizations are now starting to leverage for improving their businesses. Location Intelligence (LI) is now a growing field that uses spatial information analysis and mapping tools, to derive trends and actionable insights, vital to framing better strategies in a vast amount of industries. Among other applications, Retail Site Selection can greatly benefit from modern LI implementations, with the aim of increasing sales and reducing marketing costs, by identifying the right place-to-be for any product and their target audience. In this presentation we will discuss some of the possibilities that Python offers in terms of geo-data visualization and analysis, and show how we can utilize machine learning algorithms within LI, to generate accurate predictions on real retail expansion case studies.TagsAnalytics, Machine-Learning, GEO and GIS
I’m a Data Scientist for Hoverture. I have a background in physics, with a PhD in quantum photonics. I studied and worked in London at UCL, and I was a researcher in Barcelona, at ICFO. Analyzing, visualizing and interpreting data it’s always been part of my research activity. Now, I have fun translating business problems into data science models!