Designing a Data Algorithm Prediction Model based on R

Authors

  • Qamber Ali Yagob Ali
  • Jun Lee
  • Hyoung-Jin Kim

DOI:

https://doi.org/10.14419/ijet.v7i3.33.21184

Keywords:

Splitting data, utility functions, Cross validation, Analysis, Accuracy, Visualization, Prediction.

Abstract

This study is about data analysis and prediction model using R open source language, R is a language and environment for statistical computing and graphics so it’s a full function programming language. In this study through using the Titanic datasets, we created a model that predict the survival rates of the test dataset, we used the Train dataset that has  the survived variable with levels of "0"(perished) and "1"(survived) data and the test dataset with no survived levels to predict the survival rates on the test. In this study, we used the R functions, packages, and machine learning algorithms that provided in R, to combine, analysis and splitting the data we created utility functions to make features and predictive potential values for our new variables. We Analyzed the Training data set by cross-validation with 82% accuracy, visualization, and decision tree and then leveraged it to test data to predict the survival rate on test data.

 

 


 

References

View Full Article:

How to Cite

Ali Yagob Ali, Q., Lee, J., & Kim, H.-J. (2018). Designing a Data Algorithm Prediction Model based on R. International Journal of Engineering & Technology, 7(3.33), 279–284. https://doi.org/10.14419/ijet.v7i3.33.21184
Received 2018-10-07
Accepted 2018-10-07