PJI_analysis

These are the projects in Data Mining course in NYCU.

introduction

There are two projects in this repository

For Project1

Professor asked us to use the [1] by J. Quinlan to be the classifier in project1.
The goal of project1 is to do data preprocessing and use c45 source code to classifiy the datapoints
In this project, since source code of [1] is extremely old,
you need to modify the source code to fit your os and complier
For example, change "cfree" to "free"
Or change "exit" to "exit(1)", and so on.

For Project2

Professor asked us to tried kinds of classifiers and datasets to decide which combination is the best for classification
Do a comprehensive survey on different combinations.

pdf files above are the introduction of corresponding project.

Environement

• For Project1
  • C
  • python3.7.7
  • sklearn0.24.2
  • pandas
  • ubuntu 20.04
  • GCC 9.3.0
• For Project2
  • Python3.7.7
  • sklearn0.24.2
  • pandas
  • pytorch1.9.1
  • colab with Tesla P100 GPU

reference

1. http://www2.cs.uregina.ca/~dbd/cs831/notes/ml/dtrees/c4.5/tutorial.html