IBM C2090-719 問題集

質問 1：
You manage a store and want to increase the sales of some high profit items. One way to do this is by placing these items close to other products that are often purchased with the high profit items.
You have a year's history of all sales, including all the items that were purchased in each sale. To determine optimal product placement for the high profit items, which mining method should be used to analyze your data?
A. Classification
B. Regression
C. Association
D. Clustering
正解：C

質問 2：
Which two functions can the SQW tool perform for a data flow? (Choose two.)
A. parallel process execution
B. operators for row compression, roll-in, and roll-out
C. business processes such as secure command and secure FTP
D. range partitioning for staging tables
E. filtering and sorting for operator properties such as table columns
正解：D,E

質問 3：
What are two primary goals of denormalization? (Choose two.)
A. to simplify the querying of the database
B. to decrease the redundancy of data within a database
C. to increase the number of tables in the model
D. to allow for data compression
E. to improve performance of a system
正解：A,E

質問 4：
When connecting an Output Port of a Table Source to the Names Port of a Sequences Operator, what are two valid methods of mapping columns? (Choose two.)
A. by Index
B. by Position
C. by Name
D. by Key
E. by Attribute
正解：B,C

質問 5：
As a system administrator, you have defined a cube server, specifying that all cubes are to start automatically, and added a cube to that server. You created a script to start the cube server from the command line. Wanting to make sure that your users have good performance, you want the members of your cube to be cached prior to the first user's access of the cube. What will allow you to accomplish this?
A. Add to the cube server startup script a member seed query for the cube.
B. Use the Admin Console and specify the execution of an MDX seed query when the cube starts.
C. Use the Admin Console and specify the execution of a member seed query when the cube starts.
D. Just execute the cube server startup script.
正解：D

質問 6：
Which tool should be used to design a mining flow?
A. Use the data model editor to build the mining data mart.
B. Use the mining editor to analyze existing data for the mining specific tasks in a production database.
C. Use the data flow editor to build the mining data mart.
D. Use the data flow editor to populate a new data mart from normalized tables in your warehouse or from external sources like flat files.
正解：B

質問 7：
Which statement describes the conceptual data model?
A. The conceptual data model is a high-level business view of the data.
B. The conceptual data model is where the data model actually gets defined.
C. The conceptual data model contains the configuration, keys, indexes selected, and constraints.
D. The conceptual data model includes the DDL file for the warehouse.
正解：A

質問 8：
Several operators are defined and linked together in DataFlow1. Another set of operators make up DataFlow2. A control flow is defined and both DataFlow1 and DataFlow2 areused.
You require that DataFlow1 dynamically change the variable values used in DataFlow2.
How can you fulfill this requirement?
A. Using the File Export operator, DataFlow1 writes a file, containing updated variable values. A variable assignment operator is then used to assign the values in the file to the appropriate variables. DataFlow2 then has access to the updated variable values.
B. The inherent design of the SQL Warehouse Tool is that any variable value changed in one data flow is accessible by any other data flow as long as the data flows are defined in the same warehouse project.
C. When a control flow is executed, a run profile provides the initial values for all variables.
Once those values are set in the run profile, they are in affect for the entire execution of the control flow.
D. Using the File Export operator, DataFlow1 writes a file that contains updated variable values. DataFlow2 accesses those updated variable values by reading that same file using an Import File operator.
正解：A

質問 9：
What correctly describes how the InfoSphere Intelligent Miner components are used for predictive modeling?
A. The scoring component is first invoked to score an existing model to generate a new predicted value. This value is then used by the modeling component to generate a new predictive model.
B. The visualization component can only be used to view the scoring results. The model can beviewed by the modeling component.
C. The modeling component creates a new mining model based on existing data. The scoring component then trains and applies this model to new data in a DB2 table to generate a predicted value.
D. The modeling process is first invoked to create and train the prediction model based on existing data. The scoring process then applies this model to new data to generate a predicted value.
正解：D