title: “R Notebook” output: html_notebook author: Adhislacy
This is an R Markdown Notebook.
**PROJECT PROPOSAL.*
DATA SET : BLOOD TRANSFUSION SERVICE CENTER.
- The data was collected in from Hsin-Chu City in Taiwan, during a university visit for the blood donation. There are 748 instances and 5 attributes, ranging from; how recent the subject had donated, the frequency, among other things.
- Location of the data : https://archive.ics.uci.edu/ml/machine-learning-databases/blood-transfusion/
- Type of algorithm to be used: Decision trees (C5.0). This is because I want to run the classification, and to also find out what attributes are help in classification.
PLAN.
- Since the data is already collected, we shall go ahead and feed it in R and run the desired algorithm(s).
- Prepare and explore the data If required, I will clean the data by eliminating unnecessary data. Look at the donors’ charactristics, class variable. Create a rndom sample for training and test data. Split the data frames. Check the proportion of the class variable(s).
- Train and test the model Build decision trees. Display simple facts about the tree. Display detailed information about the tree.
- Evaluate model performance. Create a factor vector of predictions on test data. Create a cross tabu,ation of predicted vs actual classes.
- Improving the model performance. Boosting decision trees’ accuracy (with N numner of trials). Create dimensions for donor matrix. Build the matrix. Apply the matrix to the tree.
- Give a final report of the output/results.