• Primary Tumor Data Set:https://archive.ics.uci.edu/ml/datasets/Primary+Tumor
• The data provides 18 columns (note: all values are numeric and correspond to their index, see the description on the dataset page). See the file `dataset.names` for the names of the fieldTask
• For every combination of features, fit a linear boundary (classifier).
• Use this linear boundary to predict the same data.
• Compute the accuracy, for e.g

X	Y (label)	Accuracy
Class	Age	0.617
Class	Skin	0.938
Age	Neck	0.870

• You will have 18 * 17 (306) rows in your table
• Some rows for some columns havemissing values or special characters like "?". Fill those in with what you think is appropriate. But it is important to generate results on all combinations (see the column 'Sex' for what I meant). Google 'data imputation techniques' and go from there.

s.