Notice: This page requires JavaScript to function properly.
Please enable JavaScript in your browser settings or update your browser.Lära Setting Parameters: Affinity | Spectral Clustering
Cluster Analysis in Python
course content

Kursinnehåll

Cluster Analysis in Python

Cluster Analysis in Python

1. K-Means Algorithm
Exploratory Data AnalysisWhat is a K-Means Algorithm?Defining the Number of ClustersCase 1: Three Distinct ClustersCase 2: Four Distinct ClustersClustering Weather DataFebruary vs July Average TemperaturesVisualizing the Dynamics Across Clusters
2. K-Medoids Algorithm
What is a K-Medoids Algorithm?Cluster CentersSilhouette ScoresK-Medoids and the Weather DataMean Yearly Temperatures Across ClustersComparing the Dynamics
3. Hierarchical Clustering
What is a Hierarchical Clustering?DendrogramsSetting Parameters: LinkageRand IndexThe Weather Data and LinkagesDeciding the Number of ClustersWeather Data: Complete and Ward LinkagesHow Similar are the Results?
4. Spectral Clustering
What is a Spectral Clustering?Peculiarity of Spectral ClusteringSetting Parameters: AffinityDefining the Number of ClustersIs 4 the Optimal Number of Clusters?Comparing the Trends Across Clusters

book
Setting Parameters: Affinity

Well, that was not the result we were looking for. Can we improve it? Can we make the clustering algorithm learn to differ such structures?

The answer is yes - we need to set some parameters within the SpectralClustering function. The parameter we should change is affinity. This parameter defines how should affinity matrix be built (the math explanation of this is outside the scope of this course). By default, the parameter's value is 'rbf'. If we want to differ the clusters with such a structure as in the previous chapter, we should consider the 'nearest_neighbors' value of the parameter.

Uppgift

Swipe to start coding

  1. Import SpectralClustering function from sklearn.cluster.
  2. Create a SpectralClustering model object with 4 clusters and set the affinity parameter to 'nearest_neighbors'.
  3. Fit the data to the model and predict the labels. Save predicted labels as the 'prediction' column of data.
  4. Build the seaborn scatter plot with 'x' column of data on the x-axis, 'y' column of data on the y-axis for each value of 'prediction'. Then, display the plot.

Lösning

Switch to desktopByt till skrivbordet för praktisk övningFortsätt där du är med ett av alternativen nedan
Var allt tydligt?

Hur kan vi förbättra det?

Tack för dina kommentarer!

Avsnitt 4. Kapitel 3
toggle bottom row

book
Setting Parameters: Affinity

Well, that was not the result we were looking for. Can we improve it? Can we make the clustering algorithm learn to differ such structures?

The answer is yes - we need to set some parameters within the SpectralClustering function. The parameter we should change is affinity. This parameter defines how should affinity matrix be built (the math explanation of this is outside the scope of this course). By default, the parameter's value is 'rbf'. If we want to differ the clusters with such a structure as in the previous chapter, we should consider the 'nearest_neighbors' value of the parameter.

Uppgift

Swipe to start coding

  1. Import SpectralClustering function from sklearn.cluster.
  2. Create a SpectralClustering model object with 4 clusters and set the affinity parameter to 'nearest_neighbors'.
  3. Fit the data to the model and predict the labels. Save predicted labels as the 'prediction' column of data.
  4. Build the seaborn scatter plot with 'x' column of data on the x-axis, 'y' column of data on the y-axis for each value of 'prediction'. Then, display the plot.

Lösning

Switch to desktopByt till skrivbordet för praktisk övningFortsätt där du är med ett av alternativen nedan
Var allt tydligt?

Hur kan vi förbättra det?

Tack för dina kommentarer!

Avsnitt 4. Kapitel 3
Switch to desktopByt till skrivbordet för praktisk övningFortsätt där du är med ett av alternativen nedan
Vi beklagar att något gick fel. Vad hände?
some-alt