Dimensionality Reduction & Spectral Techniques

  • 8 videos | 40m 34s
  • Includes Assessment
How do we get from raw data to improving the level of performance? The answer is found in this opening course. This course will introduce us to the tools and techniques developed to make sense of unstructured data and discover hidden patterns.


  • Understand what networks are and how to analyze them
    Be able to use pca
    Know how to use eigenvectors and the covariance matrix
    Understand how clustering occurs in graphs and networks
  • Know how eigenvectors can be used to capture the connectivity structure of a large network
    Know how to use the eigenvectors of the laplacian matrix to find meaningful clusters that respect hidden structure in the data
    Understand modularity clustering and how it works
    Know what embeddings are and understand their uses


  • 4m 7s
    Learn about tools that tell us a whole lot about data. FREE ACCESS
  • 5m 11s
    Now learn about finding major patterns in data using principal component analysis. FREE ACCESS
  • Locked
    3.  The Magic of Eigenvectors 1
    4m 50s
    You just learned about PCA. Now learn how to compute it. FREE ACCESS
  • Locked
    4.  Clustering In Graphs And Networks
    4m 33s
    Learn how clustering occurs in graphs and networks. FREE ACCESS
  • Locked
    5.  The Magic of Eigenvectors 2
    5m 16s
    See how useful eigenvectors can be when describing the connectivity structure of a large network. FREE ACCESS
  • Locked
    6.  Spectral Clustering
    5m 30s
    In the past videos we have seen criteria for finding communities and we've seen that eigenvectors capture important properties of the network. Now we'll put everything together. FREE ACCESS
  • Locked
    7.  Modularity Clustering
    5m 23s
    In the previous video you saw how to explicitly use eigenvectors to recover hidden communities in a graph. Now learn new criterion that automatically determines the number of communities. FREE ACCESS
  • Locked
    8.  Embeddings And Components
    5m 45s
    All the methods learned so far involve new feature vectors for the data points. These are know as embeddings. Learn about different types and their uses. FREE ACCESS


Rating 5.0 of 1 users Rating 5.0 of 1 users (1)
Rating 5.0 of 1 users Rating 5.0 of 1 users (1)