2. Data Mining
Objectives
Objective 1: Predicting profitability of future products for
Blackwell Electronics using two different data analytics
methods
Complete task with Similarity Analysis
Complete task with Regression Analysis
Objective 2: Predict which brand of computer Blackwell
Electronics customers prefer
Use both actual and predicted data to solve the problem
Use various methods to produce a high accuracylow risk result
3. Objective 1:
The difference
between
similarity and
regression
Similarity Analysis
We compare multiple lists of
numbers to evaluate their
similarity. To do this we
measure products and their
variables on a sliding scale.
The idea is similar objects
sell roughly the same.
Regression Analysis
Estimate the relationships
among variables to forecast
profitability of future products.
This is different than a
Similarity Analysis because it’s
a continuous prediction rather
than a classification.
4. Recommended
New Products
Method
Comparison
and Results
Similarity Analysis
$106,666.64
$0.00
$20,000.00
$40,000.00
$60,000.00
$80,000.00
$100,000.00
$120,000.00
Total Profit Gained
Product
Regression Analysis
$186,602.40
$0.00
$20,000.00
$40,000.00
$60,000.00
$80,000.00
$100,000.00
$120,000.00
$140,000.00
$160,000.00
$180,000.00
$200,000.00
Total Profit Gained
Product
5. Objective 2:
Brand
Preference
Report
How we did this
Investigated customer responses to survey
questions
Using data analytics methods, the goal was to
discover similarities between the data
Used data analysis to recover 5000 missing
survey entries
Combined 15,000 sets of data with the
outcome having an accuracy rating of almost
85%
7. Final Report
Recommendation
Customers prefer
Sony Computers
to Acer Computers
by a margin of
nearly 2 to 1
Pursue a deeper
strategic
relationship with
Sony
8. Summary of
Lessons
Learned for
Blackwell
Electronics
Decided top five products to introduce into Blackwell’s
inventory
Successfully predicted profits of said new products
Demonstrated how Data Analytics method used nearly
doubles profits
Predicting customer brand preference
Increase customer satisfaction to help build loyalty
Improve relationships with vendors based on product
marketability
9. Other Possible Uses for Data Analytics
Optimize inventory to eliminate products that don’t sell
Use click stream analysis to see what the customer is
looking at and how they arrived there
Understand how the customer uses the website to
help personalize the experience
Could use Sentiment Analysis in social media to
gather opinions and suggestions concerning Blackwell
Electronics
Use consumer data to extend credit to low risk
applicants