Traditional methods for predicting crop yield are often

The gap

Evidence profile

Sourced from the stated research gap and limitations and future work of the source papers, classified as general, drawn from work published between 2024 and 2026, spanning 4 journals. Those papers have been cited 59 times in total.

Supporting evidence — 4 representative gaps

• Application of crop growth models in crop yield assessment (2026) · Frontiers in Plant Science · cited 1× · doi

  The paper identifies a gap in the integration of machine learning and deep learning techniques with crop growth models. There is a need for region-specific adaptation strategies. The paper highlights the importance of considering local climatic conditions and production systems.

  generalstated research gap
  Keywords: paper identifies gap integration machine learning deep techniques
• Smart agriculture: utilizing machine learning and deep learning for drought stress identification in crops (2024) · Scientific Reports · cited 58× · doi

  38-2023 ResGCNet 39-2024 Random forest, RNA The 1D-ResGC-Net CNN model effectively detected drought stress in tomato plants, outperforming PLSDA and RF models with fewer input features. The 1D-ResGC-Net’s accuracy significantly drops with fewer input features The study shows that a location-based ML classifier effectively detects drought stress in soil metagenomes with strong generalization The focus on soil metagenomes may not cover variability in drought responses across different plants or environments 40-2021 CNN, DCNN, GoogLeNet DCNNs, like GoogLeNet efficiently detects crop water stress, allowing for real-time detection in agriculture.

  generallimitationsevidence 5/5
  Keywords: drought stress googlenet resgc effectively plants fewer input features detects soil metagenomes crop resgcnet random
• Development of an Improved Hybrid Deep Learning Model for Cassava Yield Prediction (2026) · International Journal of Mathematics And Computer Research · doi

  This study presented a novel hybrid cassava yield prediction framework that combines Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM), and Harris Hawk Optimisation (HHO). By combining CNN's spatial feature extraction capability, LSTM's temporal dependency learning strength, and HHO's adaptive hyperparameter optimisation mechanism, the proposed HHO-CNN-LSTM model effectively addresses the complexity and nonlinearity inherent in agricultural yield prediction. To ensure the experimental results' reliability and robustness, primary soil from SoilGrids250m version 2.0, meteorological datasets from the Nigerian Meteorological Agency (NIMET), and cassava yield data the Food and Agriculture Organisation Statistics were preprocessed through noise removal, normalisation, and rigorous validation using a 5- fold cross-validation approach. Using MAE, MSE, RMSE, MAPE, and R-square performance evaluation metrics, the optimised consistently outperformed the conventional CNN-LSTM model. HHObased hyperparameter tuning improves predictive accuracy, reduces error margins, and improves model generalisation. These findings show the effectiveness of metaheuristicdriven deep learning optimisation for crop yield prediction, particularly in the Nigerian agricultural context. Overall, the study contributes a robust and scalable decision-support HHO-CNN-LSTM model from 6640 AYOADE Olusola Bamidele1, IJMCR Volume 14 Issue 07 July 2026 “Development of an Improved Hybrid Deep Learning Model for Cassava Yield Prediction” tool for precision agriculture, offering valuable insights for farmers, policymakers, and agricultural stakeholders in terms of planning, risk mitigation, and food security improvement. Despite the promising findings of this study, several directions for future research are recommended. Future studies may incorporate remote sensing data, such as satellite-derived vegetation indices (e.g., NDVI, EVI), to improve spatial representation and yield prediction accuracy at finer resolutions. Also, future research may investigate the use of other advanced metaheuristic optimisation algorithms, such as Whale Optimisation Algorithm (WOA), Grey Wolf Optimiser (GWO), or hybrid optimisation strategies, to further evaluate and improve hyperparameter optimisation speed. Furthermore, extending the proposed HHO-CNN-LSTM framework to other crops and agro-ecological zones would help validate its generalisability and adaptability beyond cassava production in Nigeria. Finally, future research could focus on real-time yield forecasting systems that integrate the model into web-based or mobile platforms, allowing farmers and agricultural agencies to make timely decisions.

  generalfuture workevidence 5/5
  Keywords: yield optimisation lstm model prediction cassava agricultural future hybrid learning hyperparameter framework spatial proposed meteorological
• Crop Yield Prediction using Machine Learning Techniques for Smart Agriculture (2026) · International Journal of Science, Strategic Management and Technology · doi

  Traditional methods for predicting crop yield are often inefficient and cannot deal with big-scale agricultural data. The machine learning techniques of predicting yields prove to be quite lengthy, costly, and inaccurate due to uncertain climatic situations.

  generalstated research gapevidence 5/5
  Keywords: traditional methods predicting crop yield often inefficient cannot

Questions about this gap