Isa exam 2 part 2
Chapter 9 - correct answer Chapter 9
Look at diagram - correct answer look at diagram
What problems do operational data pose for BI Systems?: - correct answer -
Operational data is seldom suitable for sophisticated reporting and data mining
What are these problems with operational data specifically? - correct answer -Dirty
Data (putting in B as gender or for age: 213)
-wrong granularity (too fine, not fine enough)
-too much data (too many attributes, too many data points)
-Inconsistant data: a customers phone number changes, so the old data won't match
what they have on record for the new data...
Granularity: - correct answer -referring to the amount of detail represented about the
data
--too fine or too coarse
-Example (I'm searching for an app and i know what I need, but theres so many options
that I can't find it and sort through it all)
How does granularity move when you move from the top of the pyramid to the bottom? -
correct answer as a ceo you want the coarse data.. that's the overall gist of the
departments/or regions. If im a manager I want more fine data to see how each store is
doing. But in general it is better to be more fine, or at least have accesss to it. As a ceo I
want cparese data, but understand that when u get coarse you aggregate and small
detail is lost... so make sure when I get the coarse, I have acess to the finer data...
What are some of the hurdles and costs associated with "cleaning" corporate data and
why do companies do this? - correct answer -Costs very much (b/c of ppl and
software).
-Hurdles (Redundancy, Homonoyms and synomnyms, inconsistances, invalid or
inaccurate information)
Data warehouse: - correct answer -where an organizations data is managed.
Functions: Obtain, organize, clean, and relate.
Data Mart: - correct answer Is a smaller data collection than the data warehouse that
address the needs for a particular department.
-so what this is saying is that if the data warehouse is the supply chain, then the data
mart is the retail store.
-the warehouses are data management experts that know how to clean and organize it,
but the mart is where it runs its specific departmental function.
Chapter 9 - correct answer Chapter 9
Look at diagram - correct answer look at diagram
What problems do operational data pose for BI Systems?: - correct answer -
Operational data is seldom suitable for sophisticated reporting and data mining
What are these problems with operational data specifically? - correct answer -Dirty
Data (putting in B as gender or for age: 213)
-wrong granularity (too fine, not fine enough)
-too much data (too many attributes, too many data points)
-Inconsistant data: a customers phone number changes, so the old data won't match
what they have on record for the new data...
Granularity: - correct answer -referring to the amount of detail represented about the
data
--too fine or too coarse
-Example (I'm searching for an app and i know what I need, but theres so many options
that I can't find it and sort through it all)
How does granularity move when you move from the top of the pyramid to the bottom? -
correct answer as a ceo you want the coarse data.. that's the overall gist of the
departments/or regions. If im a manager I want more fine data to see how each store is
doing. But in general it is better to be more fine, or at least have accesss to it. As a ceo I
want cparese data, but understand that when u get coarse you aggregate and small
detail is lost... so make sure when I get the coarse, I have acess to the finer data...
What are some of the hurdles and costs associated with "cleaning" corporate data and
why do companies do this? - correct answer -Costs very much (b/c of ppl and
software).
-Hurdles (Redundancy, Homonoyms and synomnyms, inconsistances, invalid or
inaccurate information)
Data warehouse: - correct answer -where an organizations data is managed.
Functions: Obtain, organize, clean, and relate.
Data Mart: - correct answer Is a smaller data collection than the data warehouse that
address the needs for a particular department.
-so what this is saying is that if the data warehouse is the supply chain, then the data
mart is the retail store.
-the warehouses are data management experts that know how to clean and organize it,
but the mart is where it runs its specific departmental function.
