How can we help?

You can also find more resources in our Help Center.

39 terms

Ch.4 & Ch. 5 Case Studies

STUDY
PLAY
having trouble forecasting product demand
Problem for case study DATA MINING GOES TO HOLLYWOOD
classification (flop to blockbuster), neutral network, decision tree, SVM, 3 ensembles
Solution for case study DATA MINING GOES TO HOLLYWOOD
-making right decisions during the making of a motion picture in order to manage large amounts of money and get the highest ROI
-help determine how much to invest in the production of a motion picture
-evaluate tradeoffs to maximize success of movie production
Results for case study DATA MINING GOES TO HOLLYWOOD
to stay competitive, they needed to make decisions in real-time to increase retention, reduce costs, and maintain repeat customers
Problem for case study BUSINESS ANALYTICS AND DATA MINING HELP 1-800-FLOWERS
SAS data mining tools to discover novel patterns about its customers and turn that knowledge into business transactions.
Solution for case study BUSINESS ANALYTICS AND DATA MINING HELP 1-800-FLOWERS
-more efficient marketing campaign
-decreased mailing, increased responses
-better customer experience
-increase repeat sales
Results for case study BUSINESS ANALYTICS AND DATA MINING HELP 1-800-FLOWERS
determine a way to quickly and easily find patterns and trends in unsolved criminal cases
Problem for case study POLICE DEPARTMENT FIGHTS CRIME WITH DATA MINING
Kohonen networks to cluster similar cases within SPSS software
Solution for case study POLICE DEPARTMENT FIGHTS CRIME WITH DATA MINING
able to reopen unsolved cases
Results for case study POLICE DEPARTMENT FIGHTS CRIME WITH DATA MINING
Needed a better way to detect crimes such as customs fraud, income tax evasion, money laundering, and terrorist funding.
Problem for case study A MINE ON TERRORIST FUNDING
Used data mining techniques to analyze financial transactions and detect criminal activity.
Solution for case study A MINE ON TERRORIST FUNDING
More efficient evaluation of financial transaction data aided in fighting terrorism and increased the quality of intelligence information.
Results for case study A MINE ON TERRORIST FUNDING
predict the survivability of prostate cancer
Problem for case study DATA MINING IN CANCER RESEARCH
3 data mining techniques in conjunction with logistic regression (classification analyses)
Solution for case study DATA MINING IN CANCER RESEARCH
data mining techniques can be used to develop models that posses a high degree of predictive and explanatory power
Results for case study DATA MINING IN CANCER RESEARCH
Find un- or misdiagnosed patients with illnesses that qualify for higher Medicare reimbursements
Problem for case study HIGHMARK INC., EMPLOYS DATA MINING TO MANAGE INSURANCE COSTS (maximize medicare revenues)
SAS Enterprise Miner to build decision trees that map likely outcomes a patient faces based on measures such as symptoms, health history and demographics.
Solution for case study HIGHMARK INC., EMPLOYS DATA MINING TO MANAGE INSURANCE COSTS (maximize medicare revenues)
The insurer estimates it saves millions by finding patients with one of 27 illnesses that qualify for higher Medicare reimbursements
Results for case study HIGHMARK INC., EMPLOYS DATA MINING TO MANAGE INSURANCE COSTS (maximize medicare revenues)
customers today face many varieties of beers from which to choose
Problem for case study COORS IMPROVES BEER FLAVORS WITH NEURAL NETWORKS
used neural networks to link chemical composition to sensory analysis
Solution for case study COORS IMPROVES BEER FLAVORS WITH NEURAL NETWORKS
time gained from disposing of panel tests for flavor
Results for case study COORS IMPROVES BEER FLAVORS WITH NEURAL NETWORKS
sought to identify best predictive modeling technique to help manage customer churn in the wireless telecommunications industry
Problem for case study PREDICTING CUSTOMER CHURN - A COMPETITION OF DIFFERENT TOOLS
churn scorecards so retention effort is focused on the relevant customer
Solution for case study PREDICTING CUSTOMER CHURN - A COMPETITION OF DIFFERENT TOOLS
increased customer retention
Results for case study PREDICTING CUSTOMER CHURN - A COMPETITION OF DIFFERENT TOOLS
needed a comprehensive tool in order to offer a comprehensive set of intelligence services
Problem for case study DATA MINING HELPS DEVELOP CUSTOM-TAILORED PRODUCT PORTFOLIOS FOR TELECOMMUNICATION COMPANIES
Statistica Data Miner
Solution for case study DATA MINING HELPS DEVELOP CUSTOM-TAILORED PRODUCT PORTFOLIOS FOR TELECOMMUNICATION COMPANIES
attractive per-minute calling rates
Results for case study DATA MINING HELPS DEVELOP CUSTOM-TAILORED PRODUCT PORTFOLIOS FOR TELECOMMUNICATION COMPANIES
need to analyze news from several sources
Problem for case study MINING TEXT FOR SECURITY AND COUNTERTERRORISM
The Genoa Project, part of the DARPA's total information awareness program
Solution for case study MINING TEXT FOR SECURITY AND COUNTERTERRORISM
-reduced response time
-increase in the probability of taking the best action
-quickly detect patterns in the form of actionable information
Results for case study MINING TEXT FOR SECURITY AND COUNTERTERRORISM
nearly impossible to efficiently process enormous amounts of semi-structured data
Problem for case study TEXT MINING FOR PATENT ANALYSIS
text mining software tools
Solution for case study TEXT MINING FOR PATENT ANALYSIS
-enable competitive intelligence
-make critical business decisions
-recruit best talents
-identify unauthorized use of patents
-prevent competitors from inventing similar products
Results for case study TEXT MINING FOR PATENT ANALYSIS
-combining data from structured databases with unstructured data from text
-customizing the software package's vocabulary for words used
-bringing in text data that is not in traditional data warehouses
Problem for case study HP AND TEXT MINING
specific practical application of text mining to extract useful information from the thousands of customer communications
Solution for case study HP AND TEXT MINING
-found insights into HP's customers
-the methods used are generally applicable across all such industries
Results for case study HP AND TEXT MINING
Prevent fraudulent healthcare insurance claims from getting paid.
Problem for case study HIGHMARK INC., EMPLOYS DATA MINING TO MANAGE INSURANCE COSTS (healthcare fraud prevention)
SAS Enterprise Miner automates modeling to make it easier for investigators to spot questionable activity.
Solution for case study HIGHMARK INC., EMPLOYS DATA MINING TO MANAGE INSURANCE COSTS (healthcare fraud prevention)
$11.5 million in savings in 2005; work that used to take eight hours now takes minutes; investigators can handle a 30-percent increase in caseloads.
Results for case study HIGHMARK INC., EMPLOYS DATA MINING TO MANAGE INSURANCE COSTS (healthcare fraud prevention)