This study consists of a comparative analysis of various machine learning models on vertebral column dataset.
The vertebral column dataset was initially processed using appropiate pre-processsing techniques. Then the data was analysed and visualised using various machine learning models which are listed below:
- K Nearest Neighbor
- Decision Tree
- Support Vector Machine
- Extra Trees
- Random Forest
This dataset has 310 instances which represent patients belonging to 3 different classes and 6 biochemical attributes derived from shape and orientation of pelvis and lumbar spine: pelvic incidence, pelvic tilt, lumbar lordosis angle, sacral slope, pelvic radius and grade of spondylolisthesis.
The dataset can be used for Binary Classification as either Normal or Abnormal. This dataset can also be used for 3 class classification which are mentioned below:
- Normal
- Disk Hernia
- Spondylolisthesis
I have achieved an accuracy of 81% for binary classification and 89% for 3 class classification using KNN.
This dataset is available here.