ahoirg / data-visualization-tool Goto Github PK

Create a new child class of Vis which is used for the Density Plot. You can choose appropriate marks for the data (DataMarks) (e.g., sphere, cross,..) to better identify the calculated points.
Draw a line connecting all the calculated points to form a continuous distribution. You can achieve this by extending the given classes and Methods (Do not forget to add the new Vis class as option to the VisType enum).
Again check your plot/calculations to ensure the Density Plot is accurate.

The readme file should be edited.

Create a new child class of Vis which is used for the Histogram.
Draw a bar as mark for the data point (DataMark), with the height of the bar representing the frequency of the data in that bin. Add the new Vis class as option to the VisType enum and display it.
Check your calculations and binning to ensure the histogram is accurate and set correct Axis labels

Create a new child class of Vis which is used for the Violin Plot. You can choose appropriate marks for the data (e.g., rectangle, cross,..) to represent the quartiles (median, 25%, 75%), maximum and minimum.

Draw the continuous distribution to the left and right of the center of the plot (median). You can duplicate and rotate the continuous distribution generated by the KDE for that. The exact look we leave to you - but please think about why you take which design.

For example, the background of the violin plot can be colored or outliers can be marked by dots. Additional, meaningful, extensions can bring bonus points.

Tip: Complete the minimum requirements first and extend them if there is time left. Again, check your plot/calculations to ensure the Violin Plot is accurate.

For this task, you can reuse and extend the previous calculations. To generate the Violin Plot you will need the KDE calculation but also additional statistical information like the median, the interquartile range (IQR) and the extreme points in the data.

Create one (or more) classes that calculate and store these information.
Don't forget to design classes with extensibility and maintainability (OOP) in mind and comment them appropriately.

The program is needed to restart after plotting graph for new plotting in graph.
It should be returned to the main menu without the need for restart.

Creating a menu to select which graph will be draw.

Create the needed classes for the calculation of the histogram properties (OOP Paradigm!).
Choose the number of bins. For this test at least two different binning techniques. For
example you can use Sturges' formula [6]. After determining the appropriate number of bins
calculate the bin size (interval). Group the data into the appropriate bins based on their
values and count their number (frequency).

Read about kernel density estimation (KDE) and what it is used for. Select an attribute (dimension) of your data on which you want to apply the KDE method.
You can find the static method in the script “KernelDensityEstimation”. The KDE will smooth the data and show the overall shape of the distribution.
Consider the bandwidth and other parameters when calculating the visualization. You should be able to calculate points that have similarity to your histogram.