INTERNATIONAL JOURNAL OF SCIENTIFIC DEVELOPMENT AND RESEARCH International Peer Reviewed & Refereed Journals, Open Access Journal ISSN Approved Journal No: 2455-2631 | Impact factor: 8.15 | ESTD Year: 2016
open access , Peer-reviewed, and Refereed Journals, Impact factor 8.15
Dependency of accuracy on the ratio of train-test split.
Authors Name:
Ganesh Sahu
, Dr. Ram Chandra Barik
Unique Id:
IJSDR2210199
Published In:
Volume 7 Issue 10, October-2022
Abstract:
Linear Regression is a machine learning algorithm based on supervised learning. It executes a regression operation. Regression uses independent variables to model a goal prediction value. It is mostly used to determine how variables and forecasting relate to one another, and to do so we generally use dependent variables and independent variables. Here to know the accuracy we split out the data set into train and test data, and to perform that we take a standard splitting ratio of 8:2 or 7:3.This paper verifies whether the change in the ratio of the train-test dataset has a relation with the resultant accuracy of the model. To perform this here we gonna take a simple example data set and test it which different ratios multiple times and determine the change in accuracy. We are using simple linear regression in python 3.6 using Jupiter notebook and sklearn.
Keywords:
Keywords: Regression, Simple Linear Regression, least square method, Train-Test split, Sklearn, Machine learning, Matplotlib.
Cite Article:
"Dependency of accuracy on the ratio of train-test split.", International Journal of Science & Engineering Development Research (www.ijsdr.org), ISSN:2455-2631, Vol.7, Issue 10, page no.1163 - 1170, October-2022, Available :http://www.ijsdr.org/papers/IJSDR2210199.pdf
Downloads:
000337071
Publication Details:
Published Paper ID: IJSDR2210199
Registration ID:202399
Published In: Volume 7 Issue 10, October-2022
DOI (Digital Object Identifier):
Page No: 1163 - 1170
Publisher: IJSDR | www.ijsdr.org
ISSN Number: 2455-2631
Facebook Twitter Instagram LinkedIn