Machine Learning Project to determine best customers to acquire from marketing campaigns Overview By now you have gotten an introduction to arguably the most exciting area of data science – machine learning. To recap the content so far, you have:
• Gotten practical experience in exploring, cleaning, and preparing data • Learned about many supervised learning algorithms • Discovered techniques for improving model performance • Been introduced to clustering, the most common form of unsupervised learning • Gotten a taste of scalable predictive models using Spark
For the fourth project you’ll be applying your knowledge of supervised machine learning to build a classifier. Specifically, you’ll be attempting to classify whether potential customers will be persuaded to become customers of a bank. This project will fit well in your portfolio – prediction is a highly sought-after skill. Plus, you will build on your machine learning skills in the upcoming capstone project.
Concepts covered:
• Cleaning and preparing data • Exploring and visualizing data • Model selection • Improving machine learning model performance
The Dataset
The dataset includes 17 variables related to a direct marketing campaign of a Portuguese banking institution. The attributes include variables such as age, job, marital status, education, etc. Your goal is to predict if an individual will become a customer. You can find more information and download the dataset here.
Guidelines You need to produce an R or Python notebook that builds a classifier from the given dataset. Please provide explanations of each step of the process, from data exploration to the final model evaluation.
• Data Cleaning and Preparation: In this case the data is relatively clean, but you may still need some preprocessing, such as scaling. • Model Selection: Train at least two models on the dataset. Clearly indicate which metrics you used and the performance of each model. Be sure to address any imbalance in the data, as well as using an appropriate train/test data split. • Performance Optimization: Use regularization, hyperparameter tuning, or other techniques to further optimize your chosen model and/or help select the best model.
At the end of your notebook, provide a brief summary (one paragraph) of your model – what it is, what preprocessing and optimization you did, and the final accuracy (or another appropriate metric).