Tutoring/Assistance in solving this assignment.
Task 1: Selection
In this task, you need to find a set of 1,000 customers in the test set that contains as many potential
buyers of a caravan policy as possible. In other words, use your most accurate model that you
developed on the training set to select 1,000 most likely policy buyers. Instead of submitting 1,000
selected records, you should submit only a list of 1,000 integers that represent positions of the
selected records in the test data. Thus, if your selection contains the first, third, …, and the last record
in the test data set, you should submit a list of integers that looks like this:
OPMA 419 – Predictive Models in Business Analytics HASKAYNE School of Business | 3
1
3
…
4000
In other words, your submission should be a file with 1,000 indices - integers between 1 and 4,000. Put
each index on a separate line (without comma's!). Do not put any other text in the file, so your file
should have exactly 1,000 lines.
Task 2: Profit
Analyze the performance of your best model (which can/should be a different model than for task 1, as
the objective is different) and decide which records from the test set should be selected in order to
maximize the profit. As said before, each selected record costs $1, while each selected record that is a
policy buyer provides a profit of $15. This time, the number of selected records has to be determined
by you (if you submit an empty list your profit will be $0; if you submit all 4,000 cases your profit will
be about $15*240 - $4,000 = $-400 - we know that there are about 240 policy buyers in the test set).
Again, only submit the indices of the selected records (as in the previous task).
Expectations and Deliverables
The two tasks have the form of a competition: you should submit your selections by e-mail to
marco.bijvank@haskayne.ucalgary.ca. The maximal number of submissions is three (for each task) and
the best of these three submissions counts as final submission.
There are three deliverables for this project:
• The RapidMiner files associated with the best model for Task 1 and Task 2
• A short PowerPoint presentation (around 4-5 minutes) that highlights your approach that you
will need to present on Wednesday December 8 during the lecture time.
• A report that provides a detailed description how you have addressed the two individual tasks,
what approach/process you have taken, which learning algorithms you have selected to analyze
(including the parameter values), how the learning algorithms performed on your training and
validation dataset, and which learning algorithm (including parameter values) you have selected
for your final model.
DescriptionIn this final assignment, the students will demonstrate their ability to apply two ma
Path finding involves finding a path from A to B. Typically we want the path to have certain properties,such as being the shortest or to avoid going t
Develop a program to emulate a purchase transaction at a retail store. Thisprogram will have two classes, a LineItem class and a Transaction class. Th
1 Project 1 Introduction - the SeaPort Project series For this set of projects for the course, we wish to simulate some of the aspects of a number of
1 Project 2 Introduction - the SeaPort Project series For this set of projects for the course, we wish to simulate some of the aspects of a number of