You design a Business Intelligence (BI) solution by using SQL Server 2008. The solution includes a SQL Server 2008 Analysis Services (SSAS) database.
A measure group in the database contains transaction details. The transaction details include the price, volume of shares, trade type, and several other attributes of each transaction. You need to implement a data mining model that will estimate the future prices based on the existing transaction data.
Which algorithm should the data mining model use?
A.
the Microsoft Clustering algorithm
B.
the Microsoft Association algorithm
C.
the Microsoft Naive Bayes algorithm
D.
the Microsoft Neural Network algorithm
Explanation:
Tip: "based on the existing transaction" = "Neural"Microsoft Neural Network Algorithm
Microsoft Neural Network is by far the most powerful and complex algorithm. This algorithm creates classification and regression mining models by constructing a Multilayer Perceptron network of neurons. Similar to the Microsoft Decision Trees algorithm,the Microsoft Neural Network algorithm calculates probabilities for each possible state of the input attribute when given each state of the predictable attribute. You can later use these probabilities to predict an outcome of the predicted attribute, based on the input attributes. It is recommended for use when other algorithms fail to
produce meaningful results, such as those measured by a lift chart output. We often use Microsoft Neural Network as a kind of a last resort, when dealing with large and complex datasets that fail to produce meaningful results when processed using other algorithms. This algorithm can accept a data type of Discrete or Continuous as input
(Smart Business Intelligence Solutions with Microsoft SQL Server 2008, Copyright 2009 by Kevin Goff and Lynn Langit)Microsoft Neural Network
Neural networks were developed in the 1960s to model the way human neurons function. Microsoft has created the Microsoft Neural Network algorithm so we can use neural networks for such mundane activities as predicting product sales. Of course, predicting product sales might not seem so mundane if your future employment is dependent on being correct.
Function
The Microsoft Neural Network algorithm creates a web of nodes that connect inputs derived from attribute values to a final output. The combination function determines how to combine the inputs coming into the node. Certain inputs might get more weight than others when it comes to affecting the output from this node.
The second function in each node is the activation function. The activation function takes input from the combination function and comes up with the output from this node to be sent to the next node in the network.
Tasks
The main purposes of the Microsoft Neural Network algorithm are
– Classification
– Regression
(McGraw-Hill – Delivering Business Intelligence with Microsoft SQL Server 2008 (2009))http://msdn.microsoft.com/en-us/library/ms174941.aspx
Microsoft Neural Network Algorithm
In SQL Server Analysis Services, the Microsoft Neural Network algorithm combines each possible state of the input attribute with each possible state of the predictable attribute, and uses the training data to calculate probabilities. You can later use these probabilities for classification or regression, and to predict an outcome of the predicted attribute, based on the input attributes.
A mining model that is constructed with the Microsoft Neural Network algorithm can contain multiple networks, depending on the number of columns that are used for both input and prediction, or that are used only for prediction. The number of networks that a single mining model contains depends on the number of states that are contained by the input columns and predictable columns that the mining model uses.
Creating Predictions
After the model has been processed, you can use the network and the weights stored within each node to make predictions. A neural network model supports regression, association, and classification analysis, Therefore, the meaning of each prediction might be different. You can also query the model itself, to review the correlations that were found and retrieve related statistics. For examples of how to create queries against a neural network model, see Querying a Neural Network Model (Analysis Services- Data Mining).
Remarks
Does not support drillthrough or data mining dimensions. This is because the structure of the nodes in the mining model does not necessarily correspond directly to the underlying data.
Does not support the creation of models in Predictive Model Markup Language (PMML) format.
Supports the use of OLAP mining models.
Does not support the creation of data mining dimension.