Price Prediction with Bidirectional Long Short-Term Memory Algorithm for High-Value Commercial Crops
Purpose – This study explored the integration of technology and develop a Crop Price Prediction to address the concerns for High-Value Commercial Crops.
Method – Price of tomato and squash datasets plus software and hardware requirements were utilized in its development. Bidirectional LSTM was used in the price prediction. Root Mean Squared Error (RMSE), and loss computation were performed as an assessment method.
Results – The computed RMSE were 2.07 and 2.04 for tomato and squash respectively. Losses that started from 0.0553 and 0.0570 for the training and test dataset respectively which both finished at 0.0128 for tomato; losses for squash started from 0.0172 and 0.0176 for the training and test dataset respectively and finished at 0.0019 for the training dataset and 0.0020 for the test dataset.
Conclusion – The use of a Bidirectional Long Short-Term Memory algorithm could predict the price for High-Value Commercial Crops based on the actual price of tomato and squash for the past five years.
Recommendations – The collection of actual data for the coming years is highly recommended to increase the prediction capability of the model. The inclusion of a multivariate dataset that directly affects price fluctuation could highly contribute to the increase of the accuracy of price prediction.
Research Implications – The capability of the developed system to predict the price for High-Value Commercial Crops serves as a guide to farmers and farm investors as to the planting calendar is concerned.
Practical Implication – The results of the study provided an initial step for a move to initiate a wider project in the area of agriculture particularly price prediction of high-value commercial crops. Also, in the absence of any system that stores price information within government offices in the province, the result of this study would prove beneficial.
This work is licensed under a Creative Commons Attribution 4.0 International License.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited.