Indian Journal of Agricultural Research

SCOPUSWeb of Science
  • Year: 2025
  • Volume: 59
  • Issue: 8

Crop Yield Prediction using Deep Learning Algorithm based on CNN-LSTM with Attention Layer and Skip Connection

  • Author:
  • H. Vijay Kalmani1,*, V. Nagaraj Dharwadkar2,**, Vijay Thapa1
  • Total Page Count: 9
  • Page Number: 1303 to 1311

1Department of Computer Science and Information Technology, Rajarambapu Institute of Technology, Shivaji UniversityKolhapur, Sakharale-415 414, Maharashtra, India

2Department of Computer Science, Central University of Karnataka, Kalaburgi-585 367, Karnataka, India

*Corresponding Authors: Vijay H. Kalmani, Department of Computer Science and Information Technology, Rajarambapu Institute of Technology, Shivaji University, Kolhapur, Sakharale-415 414, Maharashtra, India, Email: vijaykalmani@gmail.com

**Nagaraj V. Dharwadkar, Department of Computer Science, Central University of Karnataka, Kalaburgi-585 367, Karnataka, India, dharwadkarn@cuk.ac.in

Online published on 9 March, 2026.

Abstract

Accurate prediction of crop production is essential for efficient agricultural resource planning. Factors such as weather, soil moistureand temperature have a direct impact on crop yields, making precise forecasting vital.

This study presents a hybrid model that enhances crop production prediction by integrating a 1D Convolutional Neural Network (CNN) with a Long Short-Term Memory (LSTM) network and an attention layer. The model is specifically applied to wheat and rice, major crops in India. The model evolves into a CNN-LSTM hybrid, designed to improve prediction accuracy by incorporating modifications, including multi-head attention and a multiplication skip connection.

When compared with conventional methods like Support Vector Regressor, Decision Tree Regressor and Random Forest Regressor, the proposed hybrid model shows significantly better performance. It achieves a Root Mean Square Error (RMSE) of 0.017, indicating low prediction error, a Mean Absolute Error (MAE) of 0.09 and a strong correlation between predicted and actual yields, with an R2 of 0.967.

Keywords

Attention layer, CNN-LSTM, Decision tree regressor, Deep learning, Random forest regressor, Skip connection support vector regressor