This assignment is to give you some exposure to the use of neural networks for classification problems.
First donwload the following Spam data base which is in matlab data format. This database contains attributes obtained from spam emails and normal emails together with their corresponding label: "spam" or "not spam", which are coded as +1 and -1. The objective in this assignment is to create and train a neural network to identify spam on email automatically, based on the attributes obtained from the emails. I suggest you to use the neural network toolbox from matlab. To get you started fast, I wrote the following matlab script to train a feedforward neural net with this data. To give you an example of how preprocessing on the data can be used, I wrote the the following preprocessing script, which you should run before the training script, and uncomment one of the first two lines of the spam_train.m script.
Note: To understand the above scripts, please read the matlab neural network tolbox documentation.
This assignment is to give you some exposure to the use of neural networks for regression problems.
First donwload the following California Housing data base which is in matlab data format. This database contains attributes of housing complexes in California such as location, dimensions, etc, together with their corresponding price. The objective in this assignment is to create and train a neural network to do the estimation of the price of a house or apartment automatically, based on the attributes obtained from it. I suggest you to use the neural network toolbox from matlab. To get you started fast, I wrote the following matlab script to train a feedforward neural net with this data. To give you an example of how preprocessing on the data can be used, I wrote the the following preprocessing script, which you should run before the training script, and uncomment one of the first two lines of the spam_train.m script.
Note: To understand the above scripts, please read the matlab neural network tolbox documentation.
The data was obtained from the UCI Knowledge Discovery in Databases Archive . Please check that website for other data bases you could play with, or just to get more information on the Spam and California Data.