In this 1.5-hour long project, you will learn how to clean and preprocess geolocation data for clustering. You will learn how to export this data into an interactive file that can be better understood for the data. You will learn how to cluster initially with a K-Means approach, before using a more complicated density-based algorithm, DBSCAN. We will discuss how to evaluate these models, and offer improvements to DBSCAN with the introduction of HDBSCAN.
Note: This course works best for learners who are based in the North America region. We’re currently working on providing the same experience in other regions.
Clustering Geolocation Data Intelligently in Python
Coursera Project Network via Coursera
This course may be unavailable.
Overview
Syllabus
- Clustering Geolocation Data Intelligently in Python
- In this 1.5-hour long project, you will learn how to visualize geolocation data clearly and interactively using Python. You will then learn a simple but limited approach to clustering this data, using the K-Means algorithm. We will develop on this notion of clustering by moving to more advanced density-based methods, namely Density-Based Spatial Clustering of Applications with Noise, known as DBSCAN, and in order to address some of its shortcomings, the more advanced Hierarchical DBSCAN (HDBSCAN). You will also learn about a simple method of addressing outliers that may exist in a clustering problem.
Taught by
Ari Anastassiou
