This is a tool, which allows you to collect information, compare them and helps you make decisions based on them.
| Name | Description |
|---|---|
| Specimen | A specimen is a concrete object you want to compare against other specimens. Specimen have properties. |
| Property | Properties have a weight, a data type and a normalization strategy associated with them. |
| Analysis | Based on your properties, the analysis shows you which specimen best suits your defined weights. |
| Normalization Strategy | The normalization strategy declares how the values of a property get normalized. "Max" for example maps the values between maximum an 0 to 1 and 0. |
| Type | The type defines which information is held by a property. The types are: boolean, double (number) and text. Text properties are ignored for the analysis. |
| Weight | The weight of a property decides how much influence a property value has on the analysis. |
The GUI consists of three tables: Specimens, Properties and Analysis. All columns can be sorted by clicking on the column header.
The specimens table contains a row for each specimen you added. There is one column for each property you defined and an additional "name" column. You can edit specimens by double-clicking the row.
The properties table has a row for each property you defined. The columns show the name, weight, type and normalization strategy for each property. You can edit properties by double-clicking the row.
In the analysis table, you can see a row for each specimen you added and a column named "Value". The value tells you how the values and weights of the properties for each specimen scale against the other specimens.
The property weights and specimen properties are used to calculate the value in the analysis table. A very important factor is the normalization strategy, which tells the program which value for a property is "better".
After each value was scaled between 1 and 0, the values are summed up and divided by the sum of property weights. This gives a score between 1 and 0.
The strategy used to normalize these values can be set when creating the property. The following strategies are available:
- Max (red): Maps values from (min,0) to (max,1) linearly
- Min (blue): Maps values from (min,1) to (max,0) linearly
- QuartMax (green): Maps values from (min,0) to (max,1) quartically (bent towards 0)
- InverseQuartMax (black): Maps values from (min,0) to (max,1) quartically (bent towards 1)
- QuartMin (purple): Maps values from (min,1) to (max,0) quartically (bent towards 0)
- InverseQuartMin (black): Maps values from (min,1) to (max,0) quartically (bent towards 1)
- Copy/Paste
- Ctrl+Z/Ctrl+Y