Building a RapidMiner Process with Linear Regression Model

jDataLab

1 minute read

This post shows how to construct a simple predictive learning process in RapidMiner Studio by using the linear regression model to predict a continuous value.

Simple linear regression is explained at the post Linear Regression Analysis.

What is predictive learning? Read the post [Predictive Learning from an Operational Perspective]({{ref “2017-02-10-predictive-learning.md” >}})

Linear regression model explains the relationship between a quantitative label and one or more predictors(regular attributes) by fitting a linear equation to observed objects (with labels). The developed linear model will predict the label for unlabeled objects.

Launch RapidMiner Studio. Start a new process.

Download the dataset LR-dataset.csv

The dataset is made artificially from the following linear model

1y = 2 + 3x + (random Gaussian noise with sd=20)

Add the operators that are included in the following process picture. Connect the ports to enable data flows. Each operator requires specific parameter settings.

Download the Sample Process

The sample process: right click to download linear-regression-sample-rm-process.xml