Next Article in Journal
A Five-Level Converter in a Three-Level Mode for Common-Mode Leakage Current Suppression in PV-Generation Systems
Next Article in Special Issue
Artificial Intelligence-Based Decision-Making Algorithms, Internet of Things Sensing Networks, and Deep Learning-Assisted Smart Process Management in Cyber-Physical Production Systems
Previous Article in Journal
Optimization of Action Recognition Model Based on Multi-Task Learning and Boundary Gradient
Previous Article in Special Issue
Advancing Logistics 4.0 with the Implementation of a Big Data Warehouse: A Demonstration Case for the Automotive Industry
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Electronics
Volume 10
Issue 19
10.3390/electronics10192381
Review Report
electronics-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
Article
Peer-Review Record

Empowering Commercial Vehicles through Data-Driven Methodologies

Electronics 2021, 10(19), 2381; https://doi.org/10.3390/electronics10192381
by Paolo Bethaz 1,*, Sara Cavaglion 2, Sofia Cricelli 2, Elena Liore 2, Emanuele Manfredi 2, Stefano Salio 3, Andrea Regalia 2, Fabrizio Conicella 3, Salvatore Greco 1 and Tania Cerquitelli 1
Reviewer 1: Anonymous
Reviewer 2: Anonymous
Electronics 2021, 10(19), 2381; https://doi.org/10.3390/electronics10192381
Submission received: 4 September 2021 / Revised: 24 September 2021 / Accepted: 27 September 2021 / Published: 29 September 2021
(This article belongs to the Special Issue Big Data and Artificial Intelligence for Industry 4.0)

Round 1

Reviewer 1 Report

My main comment to the author is regarding the amount of data used to train the neural network (NN). The performance of neural network is very dependent on

  1. The network design
  2. the amount of data used to train the network.

Since the authors showed that there results are better than neural network, my question is also twofold

  1. Can a better neural network be designed for the application(s) in question: DTC or battery life prediction ( a different network for each case)?
  2. If we have more data say 1000x more data, will the NN outperform the TETRAPAC?

These are difficult questions. So a concrete answers may not be possible. However, a discussion on these topics will make the paper more complete. 

Author Response

We thank the reviewer for the interesting observation, which gives us the chance to better explain and clarify these important aspects.

Regarding the first question, we choose Recurrent Neural Networks because they are the most suitable for temporal data to the best of our knowledge. Indeed, DTC and battery life prediction are both temporal data. Specifically, we choose LSTMs because are the most powerful RNNs and are able to learn long time relationship. In the revised paper, we better clarify this choice by adding a discussion in Section 6.3.

Instead, regarding the second question, we agree that the NN could outperform TETRAPAC with more data. However, it is difficult to estimate the order of magnitude of data needed to make this happen. Thus, we contextualize our results with the available amount of data. In the revised paper, we added a discussion on this second aspect in Section 7.3.

Reviewer 2 Report

Please, more attention to English language.

Author Response

We thank the reviewer for the comment, which gives us the chance to make the paper clearer. In the revised paper, we corrected some typos and English formulations.

Back to TopTop