HUMAN ACTIVITIES RECOGNIZATIONUSING MACHINE LEARNING AND DATA ANALYTICS

Authors

  • Mrs.D.PUSHPA Author
  • NEERAJ NISHAD Author
  • RAMASANI MANMADHA REDDY Author
  • M VIJAYA LAXMI Author
  • MOHD SAIF Author

Keywords:

k-nearest neighbor (KNN), SGD, Random Forest, Naïve Bayes

Abstract

In recent years, smartphones have emerged as indispensable tools for recognizing 
human activities, garnering significant attention in research circles. This paper 
provides a detailed overview of various research endeavors in human activity 
recognition. Artificial Intelligence (AI) models have been meticulously developed to 
discern human activities using data sourced from the UCI online repository, 
characterized by its multivariate nature. We employed a spectrum of machine 
classification techniques including Random Forest, kNN, Neural Network, Logistic 
Regression, Stochastic Gradient Descent, and Naïve Bayes to scrutinize human 
activity patterns. Additionally, we employed feature selection techniques to reduce the 
dataset's dimensionality. Precision and Recall metrics were computed, and Confusion 
Matrices were constructed for each model. Experimental findings underscore that 
Neural Network and logistic regression models exhibit superior accuracy in human 
activity recognition compared to other classifiers like k-nearest neighbor (KNN), 
SGD, Random Forest, and Naïve Bayes, albeit at the expense of increased 
computational time and memory resources. 

Downloads

Published

11-05-2024

How to Cite

HUMAN ACTIVITIES RECOGNIZATIONUSING MACHINE LEARNING AND DATA ANALYTICS. (2024). International Journal of Engineering Research and Science & Technology, 20(2), 344-351. https://ijerst.org/index.php/ijerst/article/view/291