Database Processing for BIS Questions and Answers

Answer: C

Correct answer: non-integrated data

Non-integrated data is data that must be brought together from separate sources before it can be used - here two different DBMS products holding different parts of what one report needs. It is a structural problem, distinct from dirty data, which is wrong; inconsistent data, which disagrees between sources; and format problems, which concern how a value is written.

Answer: B

Correct answer: False

The statement is right about the electronic channels and wrong to include hand delivery. A report management system delivers electronically - e-mail, web portal, dashboard, XML web services - because its value lies in reaching the right people promptly, and any physical delivery would defeat that.

Answer: B

Correct answer: False

Reporting systems summarise the current status of business activities; predicting future ones is data mining. The confusion is understandable, because a manager reads a report in order to anticipate what will happen, but the system itself does not predict - it filters, groups and calculates, and the inference is the reader’s.

Answer: B

Correct answer: False

Operational data is generally poorly suited to BI without substantial preparation: it is dirty, has missing values, is spread across non-integrated sources, is too voluminous, and is granular in ways that do not match the analysis. That gap is exactly why data warehouses, with their extraction, cleaning and preparation programs, exist.

Answer: B

Correct answer: False

OLAP uses measures and dimensions, and it is that pair that defines an OLAP report. Associations belong to market basket analysis, a data mining technique, where they express which items are bought together. Mixing the two vocabularies is easy, but they belong to the two different branches of business intelligence.

Answer: B

Correct answer: False

The two categories are reporting and data mining. Data warehousing is not a category of BI system at all but the infrastructure that supplies both of them with clean, integrated, appropriately structured data - which is why it is discussed alongside them rather than as one of them.

Answer: D

Correct answer: measures and dimensions

An OLAP report is built from measures - the quantities being summed or averaged, such as sales - arranged against dimensions, the attributes by which they are broken down, such as product, region and time. Levels exist within a dimension, allowing drill-down, but the two things the report is made of are measures and dimensions.

Answer: C

Correct answer: Modify data

A reporting system reads: it filters, sorts, groups and calculates, and it delivers the result. It does not write back to the source, and it must not, since the data it reads belongs to operational systems whose integrity depends on transactions rather than on reports. Filtering and grouping are exactly what it does do.

Answer: B

Correct answer: False

There is no such requirement, and in practice the two are often different products chosen for different purposes: the operational DBMS for transaction throughput and integrity, the warehouse DBMS for analytical query performance over large volumes. The extraction and preparation programs bridge whatever difference exists.

Answer: B

Correct answer: False

Reading operational data is one of the three standard sources for a BI system, alongside extracts from operational databases and data purchased from vendors. What BI systems do not do is process transactions; reading what transactions have already recorded is precisely how they get their raw material.