Understanding Predictive Modeling - 3.1
to 3.3 (Shyam) correctly answered to
pass
What is modeling in the data mining process? - correct answer ✔✔Modeling is the
representation of the mined data.
What are the two main types of data model creation? - correct answer ✔✔Supervised learning
and unsupervised learning.
What is supervised learning? - correct answer ✔✔A type of data model creation where the goal
or target is known.
What is unsupervised learning? - correct answer ✔✔A type of data model creation where the
goal or target is unknown.
What is the goal of supervised learning? - correct answer ✔✔To find patterns in large datasets
when the target is known.
What is the goal of unsupervised learning? - correct answer ✔✔To find patterns in large
datasets when the target is unknown.
What is predictive modeling? - correct answer ✔✔A type of modeling that uses historical data
to make predictions about future events.
,What is descriptive modeling? - correct answer ✔✔A type of modeling that focuses on
summarizing and describing data.
What are algorithms in modeling? - correct answer ✔✔Methods or procedures used to perform
calculations and make predictions in modeling.
What is entropy in modeling? - correct answer ✔✔A measure of the uncertainty or randomness
in a dataset.
What is lift in modeling? - correct answer ✔✔A measure of the effectiveness of a predictive
model in identifying patterns.
What is data mining? - correct answer ✔✔The process of discovering patterns and extracting
useful information from large datasets.
What are the benefits of understanding modeling concepts? - correct answer ✔✔To use data in
ways that benefit organizations and customers.
What factors influence the selection of modeling methods? - correct answer ✔✔Type of data
available and goals of the modeling task.
What are some common modeling concepts? - correct answer ✔✔Supervised and unsupervised
learning, predictive and descriptive modeling, algorithms, entropy, and lift.
How do insurers use data mining? - correct answer ✔✔To make more informed underwriting,
claims, and marketing decisions.
What disciplines contribute to the modeling process? - correct answer ✔✔Statistics and
machine learning.
,What is unsupervised learning? - correct answer ✔✔Learning without a defined target.
Give an example of unsupervised learning. - correct answer ✔✔Grouping policyholders in an
insurance company.
What is a danger of unsupervised learning? - correct answer ✔✔Revealing meaningless
correlations.
What is a challenge with supervised learning? - correct answer ✔✔Requires data about the
target.
How can unsupervised learning help with supervised learning? - correct answer ✔✔Provides
information to define an appropriate target.
What is the difference between predictive and descriptive modeling? - correct answer
✔✔Predictive models estimate a target value, while descriptive models study data.
Give an example of a predictive model. - correct answer ✔✔Estimating the number of bodily
injury claims exceeding $25,000.
What can predictive models be used for? - correct answer ✔✔Determining unknown values in
the past or present.
What is the purpose of a descriptive model? - correct answer ✔✔To gain insight into data.
Give an example of a descriptive model. - correct answer ✔✔Studying similarities among
claimants with large claims.
, How can information from descriptive models be used? - correct answer ✔✔To build predictive
models.
Give an example of using descriptive models to build predictive models. - correct answer
✔✔Using data on policyholders in Texas to project large claim filings.
What is the intent of a predictive model? - correct answer ✔✔To estimate a target value.
What is the intent of a descriptive model? - correct answer ✔✔To gain insight into data.
What is the purpose of unsupervised learning? - correct answer ✔✔To identify patterns or
groupings in data.
What is the purpose of supervised learning? - correct answer ✔✔To make predictions or
classifications based on labeled data.
What is the main difference between supervised and unsupervised learning? - correct answer
✔✔Supervised learning has a defined target, while unsupervised learning does not.
What can unsupervised learning reveal? - correct answer ✔✔Meaningful patterns or
correlations in data.
What is the main challenge with supervised learning? - correct answer ✔✔The availability of
labeled data for the target.
What is an attribute? - correct answer ✔✔A characteristic or feature of a data instance.
What is an instance (example)? - correct answer ✔✔A single data point or observation in a
dataset.