Transcript Data Mining lecture slides

Data Mining
Accuracy Design Goal
ACIS 1504 - Introduction to Data Analytics & Business Intelligence
Business
Intelligence
Quality
Information
Functions
5-Component
SDLC
Info Literacy
Ethics
Design
• Spreadsheet*
• Database
Data
Analytics
Big Data*
Methods
•
•
•
•
•
Data Mining*
Text Mining*
OLAP*
Visualization*
Query
Database
Benefits 
Components 
Implement with:
* Excel
 Access
Concept Map
Data Mining
Design:
Accuracy
Mixed Cell
References
Isolate
Assumptions
Implementation
Common
Functions
Objectives
• Define Data Mining
• Explain the Accuracy spreadsheet design
goal.
• Demonstrate Excel features that support
the Accuracy design goal.
Segment A:
Data Mining
Data Mining Tools
• software that searches vast amounts of data
• uses complex statistical calculations
• outputs
•
•
•
•
Trends
Patterns
Correlations
Exceptions
Data Mining in the Insurance Industry
http://www.teradata.com/Resources/Videos/Towers-WatsonUsing-Telematics-Data-to-Mitigate-Drivers-RiskGlobally/?LangType=1033&LangSelect=true
Segment B:
Formula vs. Function
Formula vs. Function
• Formula
• created by you
• = A1 + A10
• Function
• keyword defined by Microsoft
• =SUM(A1:D1)
• Functions are more flexible than formulas
Different Results
If you delete Row 5
=SUM(C3:C5)
=B3+B4+B5+B6
=SUM(C3:C6)
Segment C:
Common Functions
Common Functions
• SUM
• AVERAGE
• MIN and MAX
• COUNT and COUNTA
Common Functions for Payroll
Common Functions for Payroll
• Open Common Calculations.xls
• Select the Payroll worksheet
Common Functions for Payroll
Add functions to calculate each of the following:
1. Gross Pay for each employee is Hourly Rate
times Hours Worked.
2. Total Payroll which is equal to Gross Pay for all
employees.
3. Total Hours Worked by all employees.
4. Average Hourly Wage.
5. Number of Employees on the payroll.
LiveScribe SmartPen
https://www.youtube.com/watch?v=U4RwtmwAJ5c
Smart Pen Sales Example
Open SmartPen Sales.xls
Calculate:
• Extended Price = Quantity x Unit Price
• Subtotal = Extended Price + Shipping
• Tax is 6% of Subtotal
• Total = Subtotal + Tax
Segment D:
Mixed Cell References
Cell Address
Reference
Changes when Copied
A1
Relative
Vertically or Horizontally
$A$1
Absolute
Never
$A1
Mixed
Vertically
A$1
Mixed
Horizontally
Relative vs. Absolute
Relative Cell References
Absolute Cell References
Cell Address
Reference
Changes when Copied
A1
Relative
Vertically or Horizontally
$A$1
Absolute
Never
$A1
Mixed
Vertically
A$1
Mixed
Horizontally
Mixed Cell References
Cell Reference Decisions
• Is the calculation entered going to be copied?
• If so, which direction?
• If it’s copied vertically, do you want the row
references to change? If it’s copied horizontally,
do you want the column references to change?
• Do you want such a change to take place?
Cell Reference Decisions
Start
Stop
Copy
Formula?
Yes
Yes
Change
Rows?
Vertical
Vertical or
Horizontal?
No
Stop
Yes
Horizontal
Change
Columns?
No
No
Enter $ before
row numbers
that should
not change
Enter $ before
column letters
that should
not change
Cell Reference Decisions
=SUM(B2:D2)
90
=SUM(B3:D3)
120
=SUM(B4:D4)
90
=SUM(B5:D5)
1. Will you copy this function?
2. If so, which direction: vertical or horizontal?
3. If you copy vertically, Excel will automatically
change all relative row references. Do you
want those row references to change?
Peter Rabbit Example
Open Common Functions workbook, select Peter Rabbit sheet.
Peter Rabbit Example – Vertical Copy
• Select the Peter Rabbit worksheet.
• As you complete these calculations,
consider if any cell references need to
be mixed.
1.
2.
3.
4.
Sum Flopsy’s three scores.
Copy this sum to Mopsy & Peter.
Add Factor A to Flopsy’s sum
Do the same for Mopsy & Peter by
copying Flopsy’s formula.
Peter Rabbit Example – Horizontal Copy
As you complete these calculations, consider
if any cell references need to be mixed.
1. Average the three values for Score 1.
2. Copy this average to Score 2 and 3.
3. Deduct Factor B from the average of Score
1.
4. Do the same for Score 2 and 3 by copying
Score 1’s formula.
MPG
Example
Open
Common
Functions
workbook,
select MPG
sheet.
1. Calculate MPG for Compact Car
2. Copy to other vehicles
3. Copy to next year
Segment E:
Accuracy
Spreadsheet Design: Accuracy
• Select the correct function or construct the
correct formula. (Choose a function over a
formula.)
• Check mixed, relative and absolute cell
references.
• Double-check all calculations.
• Know the order of operations.
• Isolate assumptions.
Order of Operations
• Parentheses
• Multiplication and Division
• Addition and Subtraction
1. 6 / 2 * 4 =
12
2. 3 + 2 * 2 – 1 =
6
3. ( 2 * 5 ) + 15 / 5 =
13
Isolate Assumptions
• Store numbers in cells
• Write equations to point to cells containing
numbers
Isolate Assumption Example
Assumption
=E2+3 is incorrect
=E2+H2 is correct
Isolate Assumptions?
How Do You Spend Your Study Time?
Have I isolated assumptions?