In this small project, you will use a free/open-source data mining tool such as the KNIME Analytics Platform (https://www.knime.com/) to build predictive models for a relatively small Customer Churn Analysis data set.

In this small project, you will use a free/open-source data mining tool such as the KNIME Analytics Platform (https://www.knime.com/) to build predictive models for a relatively small Customer Churn Analysis data set. You are to analyze the given data set (about the customer retention/attrition behavior for 1,000 customers) to develop and compare at least two prediction (i.e., classification) models. For example, you can include decision trees, neural networks, SVM, k-nearest neighbor, and/or logistic regression models in your comparison. Here are the specifics for this assignment:

Download and install KNIME from https://www.knime.com/. If you have not used KNIME before, you can quickly learn how to build predictive models using the software from any YouTube video or a quick online tutorial.

You can also use MS Excel to preprocess the data (if you need to/want to).

Download Customer ChurnData.csv data file (attached next to this file)

The data are given in comma-separated value (CSV) format. This format is the most common flat-file format that many software tools can easily open/handle (including KNIME and MS Excel)

Present your results in a well-organized professional document.

Include a cover page (with proper information about you and the assignment).

Make sure to nicely integrate figures (graphs, charts, tables, screenshots) within your textual description in a professional manner. The report should have six main sections resembling the phases of CRISP-DM (Cross-Industry Standard Process for Data Mining). Recall from Chapter 4 that CRISP-DM has the following six steps:

oStep 1: Business Understanding

oStep 2: Data Understanding

oStep 3: Data Preparation

oStep 4: Model Building

oStep 5: Testing and Evaluation

oStep 6: Deployment