Peer-Review Record
STL-ATTLSTM: Vegetable Price Forecasting Using STL and Attention Mechanism-Based LSTM
Agriculture 2020, 10(12), 612; https://doi.org/10.3390/agriculture10120612
by
Helin Yin 1
, Dong Jin 1, Yeong Hyeon Gu 1, Chang Jin Park 2, Sang Keun Han 3 and Seong Joon Yoo 1,*
, Dong Jin 1, Yeong Hyeon Gu 1, Chang Jin Park 2, Sang Keun Han 3 and Seong Joon Yoo 1,*Reviewer 1:
Anonymous
Reviewer 2:
Anonymous
Agriculture 2020, 10(12), 612; https://doi.org/10.3390/agriculture10120612
Submission received: 2 November 2020
/
Revised: 4 December 2020
/
Accepted: 5 December 2020
/
Published: 8 December 2020
Round 1
Reviewer 1 Report
This is an interesting and important paper, especially concerning the socio-economic impacts. It gives a good introduction into the topic and highlights the importance to the reader.
The conclusions and the outlook to future research seem to be a little bit short, concerning the importance of the topic.
One minor editorial comment:
Line 13. Maybe it is useful to explain the abbreviation STL-ATTLSTM already in this line.
Author Response
We appreciate your consideration of this manuscript.
Author Response File: 
Author Response.docx
Reviewer 2 Report
The authors propose an STL-ATTLSTM model using the Loess pre-processing method (STL) and an attention mechanism based on long-term memory (LSTM); this model combines various types of information to predict vegetable prices. The paper is interesting but presents some aspects that need to be improved.- At the end of the introduction it should be indicated how the document is structured for a better understanding of it.
- Section 3.1 must be better explained. The acronyms should all be explained prior to use.
- Review the formulas and explanation of each of the acronyms before use to facilitate understanding in section 3.2
- The R^2 value of each model should be indicated in the experiments section.
- The measurement of the RMSE must be indicated. How have the optimal parameters of the proposed model been obtained? This must be discussed. In addition, more parameters must be added such as input neurons or number of epochs, LSTM optimization function, etc.
Author Response
We appreciate your consideration of this manuscript.
Author Response File: Author Response.docx
