Medical Data Analysis Using Machine Learning with KNN

EasyChair Preprint 3580

Abstract

. Machine Learning has been used to develop diagnostic tools in the field of medicine for decades. Huge progress has been made in this area, however, a lot more work has yet to be done in order to make it more pertinent for real-time application in our day-to-day life. As a part of Data Mining, ML learns from previously fed data to classify and cluster relevant information. Hence, the main problems arise due to variations in the big data in the individuals and huge amounts of unorganised datasets. We have used ML to figure out various patterns in our dataset and to calculate the accuracy of this data, with hope that this serves as a stepping stone towards developing tools that can help in medical diagnosis/treatment in future. Creating an efficient diagnostic tool will help improve healthcare to a great extent. We have used a mixed dataset where an individual with any severe illness in early stages or individuals who are further along, are both present. We use libraries like seaborn to construct a detailed map of the data. The fundamental factors considered in this dataset are age, gender, region of stay and Blood groups. The main goal is to compare different data to each other and locate patterns within.

Keyphrases: Data Mining, KNN, Medical Diagnosis, Seaborn, matplotlib

@booklet{EasyChair:3580,
  author    = {Sabyasachi Mohanty and Astha Mishra and Ankur Saxena},
  title     = {Medical Data Analysis Using Machine Learning with KNN},
  howpublished = {EasyChair Preprint 3580},
  year      = {EasyChair, 2020}}