Databases Reference
In-Depth Information
Exercise 2.7
Use the HousingPriceIndexYQ.csv dataset viewed in a boundary plot to
evaluate fluctuations in housing prices. This dataset contains a “Price Index”
by state and by quarter, beginning in 1975 and ending in 2010. The index is
relative to the state's median home price in 1980 quarter 1. Thus, this data can
be used to compare rates of change between states. You cannot use the data to
compare actual home prices between states.
a. In 2010, which of the 48 state contiguous states had the largest rate of
increase in home prices from 1980? Which had the second largest?
b. In 2010, which of the 48 contiguous states had the smallest rate of increase
in home prices from 1980? Which had the second smallest?
c. Using the color encoding only, without hovering to read actual values,
determine the year and quarter when home prices in Nevada peaked.
d. Using the color encoding only, without hovering to read actual values,
determine the year and quarter when home prices in Hawaii peaked.
e. By 2010, which state had dropped more from peak? Nevada or Hawaii?
f.
In 1990 quarter 1, which state's prices were still below their 1980 level?
Summary
The first two steps in the data mining process are “initial exploration” and
“dataset preparation”. They are frequently completed in parallel.
VisMiner is a tool that supports both processes using visualizations of the
data. The visualizations are designed to preattentively present patterns. Inter-
actions with the visualizations allow the user to eliminate both rows and
columns from the dataset, resulting in a more effective and efficient data
mining process.
Where advantageous, the visualizations are synchronized. Thus, interactions
with one visualization propagate changes in another.
