Group presentation2

19
Data Mining (DM) Matthew Stanley Cynthia Denise Williams Cianti Williams

description

 

Transcript of Group presentation2

Page 1: Group presentation2

Data Mining (DM)Matthew StanleyCynthia Denise WilliamsCianti Williams

Page 2: Group presentation2

Agenda

• Background information• Real World Case 4- Applebee’s

Travelocity, and Others: Data Mining for Business Decisions

• Questions• Analysis• Final Remarks• References

Page 3: Group presentation2

Background information

DM Encompasses•Statistics•Ability of systems to learn•Artificial Neural Networks Databases•Expert Systems•Data Visualization

Page 4: Group presentation2

Background information

DM Past and Present• DM can be traced back to the late

1980’s

• Early 1990’s DM recognized as a sub-application of KDD (Knowledge Discovery Database)

• Notoriety greatly increased in the 1990 ’s• Secondary to advances in technology

• DM process continue to increase

• Will expand related to desire to collect electronic data

Page 5: Group presentation2

Background information

DM Main Goals1. Analyze2. Predict future behaviors3. Gain competitive advantages4. Find patterns

a) Associationsb) Relationships

5. Summarize6. Increase revenue, cut cost

Page 6: Group presentation2

Real World Case 4• Analyzes three companies:

Applebee’s, Travelocity, and VistaPrint uses of Data Mining

• Applebee’s: restaurant• Analyzed operations at their restaurants• Used data to calculate how much time a

customer spends in the restaurant (from time of order, to food service, to payment)

• Result: Improved customer service

Page 7: Group presentation2

Real World Case 4

• Travelocity: Online travel agency• Using text analytics software (natural language

engine) from Attensity• Identifies facts, opinions, trends, etc…

• Result: effectively identifies trends which allows the company to prevent problems or anticipate customer needs more efficiently.

• VistaPrint: online graphic design services• Improved their ability to retrieve trend

information• Installed new technology->retrieved 1% of

information• Result: Able to improve customer interaction

with the website

Page 8: Group presentation2

Advantages vs. Disadvantages Creating business data warehousing

ADVANTAGES DISADVANTAGES

• Predict Future Behaviors

• Gain Competitive • Advantages• Find Patterns• Summarize• Increase revenue,

cut cost

• Requires experience instatistics, domainknowledge

• Random fluctuations can be misinterpreted

• Privacy concerns

Page 9: Group presentation2

The Bandwagon effect Why not jump on the data mining bandwagon?

• Not for every business• Must be open minded• Need access to all phases of data for

complete picture• Individual privacy• Data integrity

Page 10: Group presentation2

Applebee’sOther uses/questions while analyzing data

• Total time to prepare meals and wait times

• Compare drink choices with sport events

• Zip codes on credit cards to create new locations• Blog content mining advertise specialties in area using Smartphone technology

Page 11: Group presentation2

Applebee’sOther uses/questions while analyzing data

Page 12: Group presentation2

YES NO

Innovative Thinking Does data mining stifle creativity?

• Encourages innovation

• Support for radical ideas

• Undo bad choices FasterUtilize technology such as a Creativity Engine

• Become too heartless

• All about numbers

Page 13: Group presentation2

Creativity Machine• Brings together libraries that were never intended to work together

• Users become infinitely flexible with the ability to transform data.

Innovative Thinking Does data mining stifle creativity? (no)

Page 14: Group presentation2

Learning Points

• Data mining continuing to grow• More art than numbers

Page 15: Group presentation2

Is this still a problem?

• Becoming more standardized• Diversified not centralized

Page 16: Group presentation2

Other examples in IT used for this case study

Amazon.com: example of company using data mining well. • Offers customized experience• Remember previous interests and display

relevant items • Displays items that are popular and related• Shows items that were commonly

purchased together

Page 17: Group presentation2

Final Remarks

• DM processes increased greatly over past ten years

• DM will expand related to desire to collect electronic data

Page 18: Group presentation2

References

Coenen, F., (2011). Data mining: past, present, future. The Knowledge Engineering Review:25 th Anniversary Issue, 26(1), 25-29. doi: 2259819321.

Mining the Blogosphere to Generate Cuisine Hotspot Maps. (2010). Journal of Digital Information Management, 8(6), 396-401. Retrieved from EBSCOhost.

Shonle, M., & Yuen, T. T. (2010). Compose & Conquer: Modularity for End-Users. ICSE:

International Conference on Software Engineering, 191-194. Retrieved from EBSCOhost.

Page 19: Group presentation2

Thank you!