Reference no: EM133192934
Assignment Task: Advanced Data Mining Techniques for Direct Marketing Campaigns
Task
Students will develop one or more Advanced Data Mining (ADM) techniques for saving the cost of a direct marketing campaign by reducing false positive (wasted call) and false negative (missed customer) predictions. Students could consider the scenario of the Assignment 1 to optimise a given marketing campaign with ADM techniques. Examples of ADM techniques such as Artificial Neural Networks and Gradient Boosting Machines will be given in related tutorials.
Method and Technology
To design a solution, students will apply ADM techniques discussed in lectures and will use related packages available on the R repository.
Using the ADM techniques, students will run individual experiments to find a solution providing the best accuracy of predicting client profiles on the bank marketing used for designing a solution to the Assignment 1.
The use of this data allows students to compare different DM techniques in terms of prediction accuracy.
Project Data
Download the Bank marketing data (.csv) required for individual experiments.
Individual Report
Each solution will be evaluated in terms of the prediction accuracy determining the efficiency of the marketing campaign. All submissions are made via BREO.
1. Apply an ADM technique to solve the Bank Marketing task presented by a benchmark data
2. Analyse problems required to develop a solution providing a high prediction accuracy on a given data set
3. In total 42% to pass
1. Identify a set of parameters which are required to be adjusted within an ADM technique in order to optimise the solution
2. Explain how the parameters of an ADM technique influence the prediction accuracy
3. Run experiments in order to verify the solution on a given data set
4. Analyse and compare the results of the experiments in a group and with the known from the literature.
5. Optionally make a 5-min video presentation and include a link to the Appendix in reports
ADM techniques and use cases developed in CoCalc/RStudio will be discussed during lectures and tutorials.
Attachment:- Advanced Data Mining Techniques.rar