AIDA 181 - Assignment 3 -
Understanding Predictive Modeling fully
solved
Supervised Learning - correct answer ✔✔If the goal of the data mining task (the target) is
known—for example, if an insurer wants to know whether auto policyholders under the age of
thirty are more likely to have an accident than those over thirty—the process is called
supervised learning. A challenge with supervised learning is that there must be data about the
target. Conducting unsupervised learning first may provide the information needed to define an
appropriate target for supervised learning.
Unsupervised Learning - correct answer ✔✔Unsupervised learning is conducted when there is
no defined target. For example, the same insurance company may simply want to know
whether policyholders fall into natural groupings. There is no intended answer to this query. A
danger of unsupervised learning is that it can reveal meaningless correlations—for example,
that policyholders named John are more likely to have accidents. Conducting unsupervised
learning first may provide the information needed to define an appropriate target for
supervised learning.
Predictive Model - correct answer ✔✔A predictive model is used to estimate a target value.
Although the value itself is unknown, the intent of the model is defined. For example, if an auto
insurer wants to know how many of its third-party bodily injury claims will exceed $25,000, it
might use a predictive model. Predictive models can also be used to determine unknown values
in the past or present.
Descriptive Model - correct answer ✔✔A descriptive model, by comparison, is used to study
data and gain insight into it. For example, an auto insurer might want to know what similarities
third-party bodily injury claimants with large claims have. Information gained from descriptive
models can be used to build predictive models.
Understanding Predictive Modeling fully
solved
Supervised Learning - correct answer ✔✔If the goal of the data mining task (the target) is
known—for example, if an insurer wants to know whether auto policyholders under the age of
thirty are more likely to have an accident than those over thirty—the process is called
supervised learning. A challenge with supervised learning is that there must be data about the
target. Conducting unsupervised learning first may provide the information needed to define an
appropriate target for supervised learning.
Unsupervised Learning - correct answer ✔✔Unsupervised learning is conducted when there is
no defined target. For example, the same insurance company may simply want to know
whether policyholders fall into natural groupings. There is no intended answer to this query. A
danger of unsupervised learning is that it can reveal meaningless correlations—for example,
that policyholders named John are more likely to have accidents. Conducting unsupervised
learning first may provide the information needed to define an appropriate target for
supervised learning.
Predictive Model - correct answer ✔✔A predictive model is used to estimate a target value.
Although the value itself is unknown, the intent of the model is defined. For example, if an auto
insurer wants to know how many of its third-party bodily injury claims will exceed $25,000, it
might use a predictive model. Predictive models can also be used to determine unknown values
in the past or present.
Descriptive Model - correct answer ✔✔A descriptive model, by comparison, is used to study
data and gain insight into it. For example, an auto insurer might want to know what similarities
third-party bodily injury claimants with large claims have. Information gained from descriptive
models can be used to build predictive models.
