association rules
Apriori association rule analysis of lung resection.
The purpose of the analysis was to identify comorbid symptoms in patients with lung cancer who required lung resection. Knowing which symptoms are co-morbid can help improve patient care and drug prescribing. The type of analysis is association rule learning, which attempts to find insights and hidden relationships in large datasets by exploring associations or frequent itemsets between variables (Han, Kamber, Pei, 2011). An example might be that a person who smokes (lhs, antecedent) is often accompanied by symptoms of cough and weakness (rhs, successor).
This project involves examining datasets using R. My analysis involved creating Apriori association rule learning to help understand how many variables occur together, which is difficult to do just by looking at a dataset.
The full report can be found in the file "Apriori Association Rules Assignment.pdf". The report shows a complete breakdown of my analysis, including problem identification, motivation, data exploration, data preparation, rules, results, and interpretation. Also includes tables and visualizations.
The R script is contained in the file "MHunfalvay_Assignment1.R". Instructions on how to use the program are included as comments in the R file. After opening the file, please read the instructions carefully before executing the code to ensure that the program works correctly.
source code
# Set working directory and read the data
setwd("/Users/melissahunfalvay/Documents/HUN/My Professional Development/Machine Learning Data 630/Assignments/Assignment 1")
# display the file names in the current working directory
dir()
#Use the read.csv comma