Menu
All articles

Data Drilling in Simlytiks

Sukanya Bala3 min read

d3VIEW’s Data Visualization Application, Simlytiks, offers a wide-variety of charts with different options for an enhanced data-mining experience. One of these chart features includes data drilling which involves navigating through multiple dimensions of data. A chart with data-drilling capabilities presents data columns in hierarchical layers allowing us to explore different levels within the same container.

Figure 1: Top layers contain information about lower layers. We drill down from top layers to view lower layers and drill back up to review top layers again.

Initial layers show a more broad summary, generally getting more detailed and specific as we navigate down the levels. Though, the layers don’t necessary have to connect in hierarchical format. These charts can showcase columns of data in any order with their drilling capacity as a way to view the correlation between columns or important aspects collectively. And, drill-down paths aren’t necessarily linear as certain paths may connect back to others because of shared data records.

Figure 2: Top levels, 1a and 1b, both share a drilled down level 2b. Second levels, 2a and 2c both share the drilled down level 3c.

Simlytiks has 4 main chart types that drill data up or down: Pivot Table, Vertical Bar Chart, Pie Chart and Tree Map. Let’s take a look at how these charts are drilled. For reference, the order in which we choose our data columns determines the hierarchy of the layers in these charts.

Drill Downs with Pivot Table

Pivot Table uses data-drilling based on its rows and hinge points. 

Figure 3: Data added as rows in this Fruits pivot table includes Climate, Fruit and Type with summary columns of Percentages for Location.

Multiple rows can be pivoted out to view or pivoted in to hide. Each added row becomes another level of data to be viewed within the whole or hidden away for summary analysis.

Figure 4: Here, we are opening the top levels for Climate (subtropical, temperate, tropical) to see which fruit reside under each category.

Drill Downs with Vertical Bar Chart

Vertical Bar Chart uses data-drilling in accordance to the X-axis when assigning multiple data columns to it.

Figure 5: Data included for the X axis to be drilled-down include Climate, Fruit, Type and Location.

We can drill down by clicking on individual bars and drill back up using the navigation at the top.

Figure 6: Here, we are drilling down from the temperate climate to see temperate-grown fruits, then drilling further down into apple types and their location.

Drill Downs with Pie Chart

Pie Chart uses data-drilling with multiple columns of data for its slices.

Figure 7: Data included for our pie columns to be drilled-down include Climate, Fruit, Type and Location.

Drill down through data by clicking on pie pieces and drill back up using the navigation at the top.

Figure 8: Here, we are drilling down from the temperate climate into grapes, flaming seedless grape type and its location.

Drill Downs with Tree Map

Tree Map uses data-drilling via its grouping factors. 

Figure 9: Data included for our tree map grouping columns include Climate, Fruit. Type and Location with the average of Percentages for the values.
Figure 10: Here, we are drilling down from the subtropical (Climate) to view the fruit as orange then further drilling down to see Navel as the type and Florida as the location.

We drill down through the map by clicking on the squares or groups and drill back up using the navigation at the top.

To learn more about Simlytiks visualizations, check out these two blog posts:

Read next

Generic DOE Data Analyzer Workflow now available in d3VIEW

When we have a dataset on hand, it is often of our interest to run a quick analysis and get some insights. The analysis usually includes making predictions …

Study of Machine Learning Parameters for Robustness

NOTE: Robustness reported in this article is being revised with corrections in derivative calculations. Updated results will be published shortly from …

Apply Simulated Annealing on a curve to find the optimum

Similar to the dataset_simulated_annealing_optimizer worker, which takes a dataset as input, we can consider the curve_simulated_annealing_optimizer …