Bi-ssas Introduction

first, what is it?

SSAS is a component used for SQL Server databases for BI, which enables you to create multidimensional databases and perform data mining operations on top of them. In this article, we mainly introduce some knowledge about SSAS data analysis. Next, let's get to know each other.

Business Intelligence provides solutions that capture data from a variety of data sources and can transform all kinds of data into the same format for storage, ultimately enabling users to quickly access interpreting data, providing effective data support for user analysis and decision making , then SSAS is the creation of multi-dimensional data sets to provide faster and more high-school data mining for data analysis.

second, the structure

SSAS also known as the mining structure, defines the data that is used to generate the mining model: it specifies the source data view, the number and type of columns, and the optional partitions that are divided into training and test sets. A single mining structure can support multiple mining models that share the same domain. This paper describes the relationship between data mining structure and data source and composing data mining model.

Working with data: source-to-structure-to-model

A mining structure in a diagram is based on a data source that contains multiple tables or views, and they are joined by the CustomerID field. A table contains information about the customer, such as geographic region, age, income, and gender, and the related nested table contains multiple lines of other relevant information for each customer, such as products that the customer has purchased. This diagram shows that multiple models can be generated from a single mining structure, and that these models can use different columns in the structure.

Model 1 uses CustomerID, income, age, and region, and filters data by region.

Model 2 uses CustomerID, income, age, and region, and filters data based on age.

Model 3 uses CustomerID, age, gender, and nested tables, and does not use filters.

Because the models above use different input columns, and two of these models also limit the data that is used in the model by applying filters, the results will vary greatly even if the models are based on the same data. Note that the CustomerID column is required in all models because it is the only available column that can be used as a case key.

through the above description: The basic architecture of a data mining structure: how to define a mining structure, how to populate it with data, and how to use it to create a model. The next article is implemented by a simple example.

Iii. Advantages and disadvantages

(a) data mining uses carefully researched statistical principles to discover patterns in your data to help you make informed decisions about complex issues. By applying data mining algorithms in Analysis Services to your data, you can anticipate trends, identify patterns, create rules and recommendations, analyze the sequence of events in complex datasets, and gain insight into new situations.

(ii) Data Mining in SQL Server 2014 is not only powerful and easy to access, but integrates with the tools many people like to use when analyzing and reporting. By looking at the links provided in this section, you can get the rich background information you need to know when you start to learn data mining.

Iv. Summary

through the above simple introduction, let us have a certain understanding of SSAS, as to its implementation and various details such as: the establishment of data sources, the establishment of data views, cubes, Multi-dimensional establishment of the next article through a simple example to explain.

Bi-ssas Introduction