Data Slicer simply slices numeric data (provided that they are integer values) into any given number of quantiles (to be chosen in the form). For example, if one has a database compiling information about individuals including their age, it may be useful to transform this field in bins of various significant ages. In turn, it will allow to contrast a given behavior with people under 20 y.old, or 45, etc. Keeping the original ages would end up in statistical uncertainties if the number of individuals is not sufficient.

 

Slicing strategies

Two slicing strategies are available:

regular

regular slicing generates quantiles which values are evenly distributed from the minimum to the maximum value.

homogeneous

homogeneous slices produce quantiles gathering roughly the same number of items.

Number of bins

Precise the wanted number of bins, and CorText Manager will devid your dataset according to the chosen slicing strategy.

Personalized intervals

Alternatively, one can simply type personalized intervals. For instance to produce 5 bins distinguishing between very negative, negative, neutral, positive and very positive sentiment polarity, you would enter: [-10:-4];[-3:-1];[0];[1:3];[4:10]