Filtration procedures¶
Filtration is the most important functionality of the system. A dataset contains many variants with a lot of associated data, and the system provides filtration tools to make effective selection of variants.
There are two types of filtration tools, and both are based on conditions applied to filtering properties of variants.
There is additional mechanism of functions which are aggregation information objects that are used in conditions like filtering properties if arguments of function is properly set.) See Filtering functions for details and reference.
Thus conditions on filtering properties and functions are base atomic object of any filtration procedure, and any filtration procedure contains then and possibly join them somehow.
Typed of filtration procedures:
Filter is just sequence of (atomic) conditions, and this sequence applies to variants one by one, in conjunctional way; so the result condition of filter is a join of (atomic) conditions by operator AND
Decision tree is more advanced filtration tool. It allows to combine atomic conditions by all logical operators AND, OR and NOT and build essentially more complex (but good enough for reading and understanding) rules of selection. However, the base atomic conditions used here are the same conditions as above.
See also¶
UX settings of filtration properties