Imagen de Portada

A Comparative Study of Predictive Analysis Using Machine Learning Techniques: Performance Evaluation of Manual and AutoML Algorithms

Guardado en:
Publicado en:International Journal of Advanced Computer Science and Applications vol. 16, no. 1 (2025)
Autor principal: PDF
Publicado:
Science and Information (SAI) Organization Limited
Materias:
Comparative studies
Data analysis
Machine learning
Algorithms
Performance evaluation
Automation
Cleaning
Predictions
Software
Accuracy
Predictive analytics
Datasets
Computer science
Regression analysis
Success
Discriminant analysis
Feature selection
Business metrics
Decision making
Tournaments & championships
Engineering
Literature reviews
Acceso en línea:Citation/Abstract
Full Text - PDF
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
Resumen:In this study, we have compared manual machine learning with automated machine learning (AutoML) to see which performs better in predictive analysis. Using data from past football matches, we tested a range of algorithms to forecast game outcomes. By exploring the data, we discovered patterns and team correlations, then cleaned and prepped the data to ensure the models had the best possible inputs. Our findings show that AutoML, especially when using logistic regression can outperform manual methods in prediction accuracy. The big advantage of AutoML is that it automates the tricky parts, like data cleaning, feature selection, and tuning model parameters, saving time and effort compared to manual approaches, which require more expertise to achieve similar results. This research highlights how AutoML can make predictive analysis easier and more accurate, providing useful insights for many fields. Future work could explore using different data types and applying these techniques to other areas to show how adaptable and powerful machine learning can be.
ISSN:2158-107X
2156-5570
DOI:10.14569/IJACSA.2025.0160102
Fuente:Advanced Technologies & Aerospace Database