ashenfad

A visualization of five clusters discovered on four fields from the Pima Indian Diabetes dataset.

Each cluster is represented by a ball. The cluster radii are proportional to the population of each cluster.

The y-axis shows the distance of each cluster to the current point (selected by the sliders). The lower a cluster's position, the nearer it is the current point. The order of the clusters on the x-axis is from nearest to furthest.

The initial point is the median for each field. Selecting a cluster will set the current point equal to the cluster's centroid.

Finally, the colors on each slider represent the closest cluster to the current point for that range.

ashenfad

A visualization of three clusters discovered on the Iris dataset.

Each cluster is represented by a ball. The cluster radii are proportional to the population of each cluster.

The y-axis shows the distance of each cluster to the current point (selected by the sliders). The lower a cluster's position, the nearer it is the current point. The order of the clusters on the x-axis is from nearest to furthest.

The initial point is the median for each field. Selecting a cluster will set the current point equal to the cluster's centroid.

Finally, the colors on each slider represent the closest cluster to the current point for that range.

ashenfad

A visualization of seven clusters discovered on the wine quality dataset.

Each cluster is represented by a ball. The cluster radii are proportional to the population of each cluster.

The y-axis shows the distance of each cluster to the current point (selected by the sliders). The lower a cluster's position, the nearer it is the current point. The order of the clusters on the x-axis is from nearest to furthest.

The initial point is the median for each field. Selecting a cluster will set the current point equal to the cluster's centroid.

Finally, the colors on each slider represent the closest cluster to the current point for that range.

ashenfad

A visualization of seven clusters discovered on the concrete compression strength dataset.

Each cluster is represented by a ball. The cluster radii are proportional to the population of each cluster.

The y-axis shows the distance of each cluster to the current point (selected by the sliders). The lower a cluster's position, the nearer it is the current point. The order of the clusters on the x-axis are from nearest to furthest.

The initial point is the median for each field. Selecting a cluster will set the current point equal to the cluster's centroid.

Finally, the colors on each slider represent the closest cluster to the current point for that range.

ashenfad

Dynamic scatterplot of the a sample from the abalone dataset, including the top 10 anomalies found with BigML's isolation forest.

Controls:

• Left click to choose X-axis.
• Right click to choose Y-axis.
• Alt + right click to choose color axis.
• Repeat click (left, right, or alt) for log scale.
• Hover over a point to see all field values.
• Click a multi-point (larger circle) to cycle through values.

ashenfad

Dynamic scatterplot of the a sample from the wine quality dataset, including four clusters found with BigML's kmeans.

Controls:

• Left click to choose X-axis.
• Right click to choose Y-axis.
• Alt + right click to choose color axis.
• Repeat click (left, right, or alt) for log scale.
• Hover over a point to see all field values.
• Click a multi-point (larger circle) to cycle through values.

ashenfad

Dynamic scatterplot of the 1985 automobiles dataset.

Controls:

• Left click to choose X-axis.
• Right click to choose Y-axis.
• Alt + right click to choose color axis.
• Repeat click (left, right, or alt) for log scale.
• Hover over a point to see all field values.
• Click a multi-point (larger circle) to cycle through values.

ashenfad

A visualization of blood plasma from the UCI diabetes dataset. The distribution is stored with a streaming histogram.

• Brush to zoom.
• Click to zoom out.
• t to toggle trimming some of the outliers from the distribution.
• r to toggle rounding populations for each bin.
• i to toggle the distribution interpolation mode.

ashenfad

A visualization of sulphates in wine using distributions from two separate clusters (kmeans). The distributions are stored with a streaming histogram.

• Brush to zoom.
• Click to zoom out.
• t to toggle trimming some of the outliers from the distribution.
• r to toggle rounding populations for each bin.
• i to toggle the distribution interpolation mode.

ashenfad

A visualization of wine densities using distributions from two separate clusters (kmeans). The distributions are stored with a streaming histogram.

• Brush to zoom.
• Click to zoom out.
• t to toggle trimming some of the outliers from the distribution.
• r to toggle rounding populations for each bin.
• i to toggle the distribution interpolation mode.