Reference no: EM133552316

Predictive Analytics

For this assignment, each group has been given a different data set. Your task is to analyse the data set using the methods you've learnt in this unit, you're more than welcome to use methods you learnt in other units in this assignment.

Data Preprocessing

This is worth 6, 10, or 14 marks depending on the difficulty of the data set. This task include, but not limited to:
• Simple statistical evaluation of the data set;
• Identify the target variable(s);
• Identify if this is a regression or a classification task;
• Are there any missing data? What can you do about them?
• Which variables are relevant or irrelevant to the task?

• What type of validation/cross-validation procedure will you use?
• Data visualisation.

Methods to use
For the analysis, you need to analyse the data using at least three different methods. You'll need to use
• Neural Networks, and
• Support Vector Machine
as two of the methods. Do not use linear and logistic regression other than as a compar- ison method. The other methods taught in this subject which you can use are:
• Ridge Lasso regression,
• Lasso regression, and
• Naive Bayes
You can substitute these methods with k Nearest Neighbours, Decision Trees, Random Forest, or other machine learning methods.
For each method, write a brief description of your steps to create the model and your prediction. What did you do? Your description should include, but not limit to, answers to the following questions:
• What is the accuracy of your model?
• Is the model a good model? Why or why not?
• Any particular choices you made or had to make in creating model or prediction? Why did you make them?

Attachment:- Predictive Analytics.rar