Predicting Multiclass classification for heart disease using supervised machine learning
This project (Predicting Multi-class classification for heart disease using supervised machine learning) is about predicting multi class classification for heart disease using supervised machine learning. Please see the section what does this document contains to know more about this project. Machine learning can be of two types.
- Supervised learning
- Unsupervised learning
Supervised learning refers to such technique which is applied to such dataset which contains target while dataset without target are trained using unsupervised learning.
Predicting Multi-class classification for heart disease using supervised machine learning
1.2) Hardware Requirement:
RAM: 2GB RAM
Operating System: Windows/Mac/Ubuntu/Linux
Python : 3.x or 2.x
Web browser: Google Chrome or Mozilla Firefox
Notebook: Jupyter (for visualizing the code on the web browser.)
Architecture: either 32 bit or 46 bit
Software: Pycharm (for running python file in easy way as well as debugging of the code)
1.3) Language Used
Python is the language used. Python files are saved with “.py” extension. Same is also implemented on Jupyter notebook whose extension is .ipynb. You need to install python as well as Jupyter for using this project. If you want to visualize the result using only python, you should install Pycharm where Jupyter notebook is used to visualize the result on the web browser.
1.4) What does this document contains:
Machine Learning is used for various purpose such as color based segmentation, predicting diseases, image processing applications such as object detection, image classification and transfer learning. When deep learning came the computation power has improved to such level that now it is possible for the machines to work like humans. Companies like Interest, Google, Facebook and Amazon is using this technology to so large extent that their revenues have increased dramatically. In this thesis we have tried to use decision tree to perform multi-class classification for heart disease.
1.4.1) What is Multi-class classification and how it is different from multi-label classification
Multi-class classification refers to having different values of target in single column whereas there are multiple categories associated with the same row/sample/data/input. Please see the next heading to download the data and look at the target or the last column. It is multi-class not multi-label problem.
1.4.2) Downloading dataset
You don’t need to worry. These data-sets are available at the UCI repository. I have downloaded from there and sharing the link directly to you. Please click on download dataset.
Please visit AI Sangam official blog article link to get anything you need to know for this project. Link is here
1.4.4) How to run this code
There are two main files Preprocessing_file.ipynb and Preprocessing_file.py where you can use Jupyter to run the first one while second can be run on the command prompt (Windows) or terminal (Ubuntu) provided python is installed. For better visualization you can use “pycharm” for running python files in the desktop or laptop.
Download the folder which will contain these files and open this path to the terminal or command prompt. I believe you have installed Jupyter notebook else please download Jupyter notebook.
Installing Jupyter on Windows:
1.) Please go to this address. I believe you have installed python on custom location which is in C drive not in program files and copy this path to the “cmd” and open this path in the command prompt. Please run command prompt as administrator.
2.) Type pip install jupyter
Installing jupyter in Ubuntu:
Open the terminal by pressing ctrl +alt + t and type pip install jupyter. Python comes pre-installed on all versions after Ubuntu 14.04. you can access it by typing the following into the terminal for Python
1.) Go to the command prompt and point to folder where you have downloaded this project folder. Copy that path and open in command window
2.) Now Type python Preprocessing_file.py
You can download “Pycharm” for both windows as well as “Ubuntu” and then import this project there and run this file. Please visit link to download Pycharm. You can either download either professional or community version.