Autoencoder vs. Regression Neural Networks for Detecting Manipulated Wine Ratings

Title data

Baumann, Michaela ; Baumann, Michael Heinrich:
Autoencoder vs. Regression Neural Networks for Detecting Manipulated Wine Ratings.
In: Mäkiö, Juho (ed.): ICCGI 2022, The Seventeenth International Multi-Conference on Computing in the Global Information Technology. - Venice, Italy : IARIA , 2022 . - pp. 7-13
ISBN 978-1-61208-972-0

Related URLs

Abstract in another language

In this study, we analyze the ability of different (neural network based) detection methods to identify manipulated wine ratings for two “vinho verde” datasets. We find that autoencoders outperform regressions in terms of true/false positive rates. All in all, neural network based autoencoders seem to detect best, while classical linear models show the smallest performance variability. Most interestingly, linear model based autoencoders perform well within a reasonable computation time. Furthermore, hyperparameter tuning via sequential accumulative selection is established.