.

ISSN 2063-5346
For urgent queries please contact : +918130348310

ERROR ANALYSIS IN IDENTIFYING FAULT DATA IN IOT DEVICES USING LASSO REGRESSION COMPARED OVER DECISION TREE MODELS

Main Article Content

Hirangkar jyoti borah, P. V. Pramila
» doi: 10.31838/ecb/2023.12.sa1.300

Abstract

Aim: The aim of this research is to perform an error analysis of fault data detection in IoT devices using lasso regression splines compared over decision tree model. Materials and Methods: Lasso regression algorithm with sample size = 20 and decision tree algorithm were evaluated to predict the efficiency percentage. Lasso regression prediction updates its weights and configurations based on the input. Results and Discussion: Lasso regression delivered significant results with 90.40% accuracy, compared to decision tree 85.80% accuracy. Lasso regression and decision tree statistical insignificance is p = 0.511 (p>0.05). Independent sample T-test value states that the results in the study are significantly not achieved with a 95% confidence level. Conclusion: Lasso regression algorithm performed significantly better than the decision tree algorithm.

Article Details