50 likes | 109 Vues
CS 204423: Assignment 1. Play around with your data. The dataset. You are given a data set named ‘Auto MPG’ The data concerns fuel consumption in miles per gallon of cars Attribute Information: 1. mpg: continuous 2. cylinders: multi-valued discrete 3. displacement: continuous
E N D
CS 204423: Assignment 1 Play around with your data
The dataset • You are given a data set named ‘Auto MPG’ • The data concerns fuel consumption in miles per gallon of cars • Attribute Information: • 1. mpg: continuous • 2. cylinders: multi-valued discrete • 3. displacement: continuous • 4. horsepower: continuous • 5. weight: continuous • 6. acceleration: continuous • 7. model year: multi-valued discrete • 8. origin: multi-valued discrete (Europe, Asia, America) • 9. car name: string (unique for each instance)
Tasks 1) Examine basic characteristics of your data 1.1 How many data points? 1.2 What is the dimensionality of the data? 2) Preprocess your data into a matrix form 2.1 How will you deal with the missing attributes? There is no absolute correct answer. Express you idea! 2.2 The last field is a string data. This can’t be stored as a matrix. What to do? 3) Find the correlations between all pairs of attributes 3.1 Take the most positively correlated variables and plot them using scatter plot. Does the result make sense? Discuss your finding. 3.2 Take the most negatively correlated variables and plot them using scatter plot. Does the result make sense? Discuss your finding. 3.3 Can you find spurious correlations? 4) Standardise your data (z-normalisation) 4.1 What is the mean of your standarised dataset? 4.2 What is the standard deviation of your standardised dataset?
Useful things • mean(), std() • max(), min() • corr() • textread() • scatter() • help ….. help textread • Give you the manual for such function
For more data sets • Visit UCI data repository • https://archive.ics.uci.edu/ml/index.html • Or collect your own