A2-JerryYe
From CS294-10 Visualization Fa07
Contents |
[edit] Auto MPG dataset
I'm using the Auto MPG dataset from the UCI machine learning collection. The Auto MPG dataset contains mileage information for several makes of American, German, and Japanese cars from 1970-1982. Several variables including cylinders, displacement, horsepower, weight, and acceleration of the vehicles are also in the dataset. The link to the dataset can be found here: http://mlearn.ics.uci.edu/databases/auto-mpg.
[edit] Setup
I used Spotfire and had to modify the data slightly into CSV format. The original dataset had some weird combination of space separated and tab separated fields. I simply replaced these with commas so that Spotfire could read the dataset.
[edit] Questions
1) Were Japanese cars more fuel efficient than American Cars between 1970 and 1982?
2) Which country made the most fuel efficient cars between 1970 and 1982?
3) How did weight contribute to fuel efficiency?
4) Did fuel efficiency improve between 1970 and 1982?
[edit] Visualizations
1) The following plot definitely shows that Japanese cars(Origin 3) were more fuel efficient than American cars(Origin 1). Also, in answering question 2, Japanese cars were the most fuel efficient, followed by German cars (Origin 2) in second, and with American cars last.
2) This visualization shows that average fuel efficiency improved for cars made in all countries during the period 1970-1982. However, there was some missing data for some countries in the late 1970's.
3) This visualization shows that there was a strong correlation between weight of the vehicle and fuel efficiency. We can easily see that as weight increased, fuel efficiency of the vehicle dropped.
