Model development is one of the steps in data mining process after preprocessing. There are several algorithms that we can use to build machine learning models. Some of the algorithms we can use include Regression, Neural Network, Decision Tree, Naive Bayes and more. This step is crucial as we are going to do predictive analytics based on this. This steps requires several tools to execute the machine learning models such as Python and RapidMiner.
For this project, me and group will build predictive models using Python and RapidMiner. There are three different experiments that we will look into after this. The experiments are as listed below:
Experiment 1 - Resampling
Experiment 2 - Resampling and Standardized
Experiment 3 - Resampling, Standardization and Discretization
As for ratio, we have two ratios that my group used for this project which are:
60:40
80:20
The algorithms that we will be testing out are:
Decision Tree
Naive Bayes
Every experiment was carried out for both ratios and both algorithims. We were trying to find the best model for data products in which I will explain on a different page.