In this series of liveProjects, you’ll learn to apply insightful graph data science techniques to real-world data problems. Making use of Python and the LynxKite graph data science platform, you’ll explore how graph data structuring can reveal new insights from highly interlinked data. Each liveProject in this series stands alone, so you can pick and choose the skills that are most relevant to you.
These projects are designed for learning purposes and are not complete, production-ready applications or solutions.
here's what's included
Project 1 Analyze the Graph Structure of Soccer
In this liveProject, you’ll construct event sequence graphs to reveal interesting information about soccer games. You’ll work to find longest pass sequences, most important players, and to understand the spatial structure of the game. You’ll define and visualize these graphs, and use connected components to find interesting event subsequences. You’ll quickly be able to uncover insights such as the most important players and the spatial structuring of the playing pitch.
Project 2 Analyze and Cluster OpenStreetMap Data
In this liveProject, you’ll use LynxKite and graph data techniques to identify some of the most important geographic points in the city of Bruges. You’ll start by downloading and processing map data, and then use a simple Python program to convert it into a graph. You’ll use graph centrality metrics to quantify the importance of vertices in your graph, and determine some of Bruges’ important locations. You’ll then use the same structure to figure out the main areas of the city without using actual district data.
Project 3 Optimize City Infrastructure
In this liveProject, you’ll use graph data optimization to determine improvements that could be made to the water infrastructure of Bruges. You’ll incorporate business data into your visualized Bruges street graph, and translate your graph into a prize collecting Steiner tree problem. You’ll then use the PCST solver in LynxKite to get an exact answer to your business problem.
Project 4 Predict Age Using GNNs
In this liveProject, you’ll build a model that can predict the age of customers for a telecom company. You’ll take basic profile data and call records for your customers, and build and improve a linear regression model using graph features. Finally, you’ll use a powerful graph neural network that can combine data from both profile features and graph structure to reliably reveal customer age range.