Multi-dimensional analysis can perform various analysis operations such as roll-up, drill-down, slicing, dicing, and rotation on the data organized in multi-dimensional form, so as to analyze the data, so that analysts and decision-makers can observe from multiple angles and multiple sides data in the database to gain insight into the information and meaning contained in the data. Multidimensional analysis fits the human mindset, reduces confusion, and reduces the likelihood of misinterpretation.
Multidimensional data analysis usually includes the following analysis methods.
1. Slicing
The selection operation performed on one dimension of a given data cube is a slice, and the result of the slice is a two-dimensional flat data. For example, in Example 2-1, selecting the usage conditions for the data cube shown in Figure 2-1: "Consignment method = On-site", "Sales Department No. = 02", "Time = 2011-01", it is equivalent to The original cube is sliced, and the results are shown in Figure 2-2.
2. Cut into pieces
A selection operation on two or more dimensions of a given data cube is a dice, and the result of the dice is a subcube, as shown in Figure 2-3.
For example, using the conditions for the data cube shown in Figure 2-1 in Example 2-1:
(Time = "March" or "April") and (Sales Department Number = "02" or "03") and (Entrustment method = "On-site")
Making a selection is equivalent to cutting a small piece out of the original cube, and the result is shown in Figure 2-4.
3. Roll up
Dimensions are hierarchical. For example, the time dimension may be composed of years, months, and days. The level of the dimension actually reflects the comprehensiveness of the data. The higher the level of the dimension, the higher the comprehensiveness of the data represented, the less the details, and the less the amount of data; the lower the level of the dimension, the lower the comprehensiveness of the data represented, the more sufficient the details, and the larger the amount of data. Roll-up, also known as data aggregation, is an aggregation operation performed in a data cube to observe more generalized data by going up in dimension levels or by eliminating a dimension or dimensions. Table 2-2 shows an example of performing a data rollup operation.
Table 2-2 Transaction volume of some sales departments in 2011 (total by year)
Sales department number Transaction volume/100 million yuan Sales department number Transaction volume/100 million yuan
01 50 03 62
02 38 04 55
4. Drill down
Drill-down, also known as data drilling, is actually the reverse operation of scrolling up, looking at the data in more detail by dropping a dimension level or by introducing a dimension or dimensions.
5. Rotate
Data from different perspectives can be obtained by pivot or rotate. The data rotation operation is equivalent to rotating the coordinate axis based on the plane data. For example, rotation may involve swapping rows and columns, or rotating one dimension into another. The result of rotating Figure 2-1 in Example 2-1 is shown in Figure 2-5.
Reprinted from http://book.2cto.com/201311/35825.html