Decision on your submission to Scientific Reports
Inbox
Ref: Submission ID bb9cb447-6a22-4579-a78c-478bdf
Dear (Dr).Ramanjaneyulu
Your manuscript "Crop Yield Prediction in Diverse Environmental Conditions Using Ensemble Learning" has now been assessed. If there are any reviewer comments on your manuscript, you can find them at the end of this email.
Regrettably, your manuscript has been rejected for publication in Scientific Reports.
Editor Comments
"The manuscript addresses a relevant application; however, it suffers from limited novelty, as the proposed approach mainly combines existing methods without sufficient methodological innovation. In addition, there are significant concerns regarding the experimental design and result consistency , which undermine the credibility of the reported performance. Therefore, the manuscript is not suitable for publication in its current form."
-Xiaojun Jin
Thank you for the opportunity to review your work. I'm sorry that we cannot be more positive on this occasion and hope you will not be deterred from submitting future work to Scientific Reports.
Kind regards,
Shruti Anand
Assistant Editor
Scientific Reports
Reviewer Comments:
Reviewer 2
The authors did some revisions, however, the literature strength was not addressed properly.
Literature survey is very weak, and should be considerably improved as suggested in the previous revision. For example, some recent metaheurictics-optimized ML approaches for agriculture were not included, that could make a stronger case for the approach taken in this study (the authors should take a look at: Evaluating the performance of metaheuristic-tuned weight agnostic neural networks for crop yield prediction; Crop Yield Forecasting Based on Echo State Network Tuned by Crayfish Optimization Algorithm; A Computer Vision-Based Approach Optimized by Modified Metaheuristic for Precise Agriculture Applications; Utilizing Modified Metaheuristic Optimizers for Computer Vision Optimization in Agriculture; A two-layer TinyML approach aided by metaheuristics optimization for leveraging agriculture 4.0 and plant disease classification).
These papers were discussed in the manuscript, however, they were not added to the reference list. Literature survey therefore, must be checked thoroughly.
Reviewer 1
accept
Reviewer 4
The novelty is limited; the AdaBoost + WOA combination is incremental and not sufficiently justified.
The methodology lacks technical depth and clear explanation.
Experimental evaluation is weak with insufficient comparison to recent state-of-the-art methods.
Reported results seem overly optimistic without strong statistical validation.
The study lacks generalizability (dataset restricted to a single region).
The manuscript shows immaturity in documentation, including poor organization, grammatical errors, and lack of scientific clarity.
Figures and analysis are mostly descriptive with limited critical insight.
Reviewer 5
1. The paper titled "Crop Yield Prediction in Diverse Environmental Conditions Using
Ensemble Learning" demonstrates limited technical novelty, with no new algorithmic or theoretical contributions, and primarily focuses on implementation and evaluation of existing methods.
2. The formula for fitness function( Eqn. 5) is inconsistent with stated objective. The defined function represents maximisation while the manuscript claims minimisation.
3. The proposed method reports a test accuracy (97%) identical to the baseline AdaBoost, raising concerns about the claimed improvement.
4. With 19698 samples and an 80:20 split, the test set should be 3940 (approx). Your confusion matrix shows only 300. Justify the selection criteria of a subset of 300 samples.
5. The inconsistency in fitness convergence curve should be removed.
6. The dataset has been described as having 9 features, whereas the SHAP analysis reports only 8. Please clarify this discrepancy.
7. The manuscript is without ablation study and performance comparison with other nature-inspired metaheuristic algorithms.
8. The reference list contains only 1 paper from 2024 and none from 2025. The authors should include recent relevant papers to ensure the work reflects the current state-of-the-art.
9. The figure captions are inconsistently placed. The authors should ensure uniform formatting throughout the manuscript.
Reviewer 6
The manuscript titled “Crop Yield Prediction in Diverse Environmental Conditions Using Ensemble Learning” addresses an important problem in precision agriculture by proposing a hybrid framework that integrates AdaBoost with the Whale Optimization Algorithm (WOA) for feature selection and hyperparameter tuning. The topic is relevant, and the use of optimization-aware ensemble learning is appropriate for handling complex agricultural datasets. The authors have made meaningful revisions in response to reviewer comments, particularly by addressing target leakage, clarifying the experimental protocol, and improving dataset transparency .
However, despite these improvements, several issues remain that should be addressed to further strengthen the manuscript.
First, the novelty of the proposed approach should be more clearly articulated. While the integration of WOA with AdaBoost is reasonable, similar optimization-based ensemble approaches have been explored in the literature. The authors should clearly explain what distinguishes their method and whether it introduces any new algorithmic contribution or improved optimization strategy beyond existing work.
Second, the reported performance, although strong, requires more careful interpretation. The model achieves high accuracy (around 96–97%) , but the manuscript does not provide sufficient evidence to fully support claims of strong generalization. The evaluation is limited to a single dataset, and no external validation is performed. The authors should either include additional validation experiments or explicitly acknowledge this limitation and moderate claims about applicability to diverse environments.
Third, the evaluation methodology could be further strengthened. While both regression and classification metrics are used, the conversion of continuous yield predictions into categorical classes may introduce bias. The authors are encouraged to provide clearer justification for this approach and emphasize metrics such as F1-score and ROC–AUC, along with detailed interpretation of confusion matrices.
Fourth, although the methodology has been improved, some sections still lack clarity. Certain equations and notations are not consistently defined, and the optimization process could be explained more concisely. Providing a simplified workflow or pseudocode would improve readability and reproducibility.
Fifth, the manuscript would benefit from minor improvements in writing and organization. Some sections remain verbose, and there are still occasional grammatical and stylistic issues. Figures and tables are relevant but should be more clearly explained and better integrated into the discussion.
In summary, the manuscript presents a promising and well-motivated approach, and the revisions made so far are appreciated. However, further improvements in novelty clarification, evaluation rigor, and clarity of presentation are required before the manuscript can be considered for publication.
Attachments:
• https://reviewer-feedback.spri
| 04:36 (6 hours ago) | |||
| ||||
Mr.D.Ramanjaneyulu N R I
Faculty of Computer Sc&Eng
Dept of Computing
Faculty of Computer Sc&Eng
Dept of Computing
