An Enhanced Deep Learning Approach in Forecasting Banana Harvest Yields

Mariannie A Rebortera,Arnel C

doi:10.14569/ijacsa.2019.0100935

Abstract

This technical quest aspired to build deep multifaceted system proficient in forecasting banana harvest yields essential for extensive planning for a sustainable production in the agriculture sector. Recently, deep-learning (DL) approach has been used as a new alternative model in forecasting. In this paper, the enhanced DL approach incorporates multiple long short term memory (LSTM) layers employed with multiple neurons in each layer, fully trained and built a state for forecasting. The enhanced model used the banana harvest yield data from agrarian reform beneficiary (ARB) cooperative of Dapco in Davao del Norte, Philippines. The model parameters such as epoch, batch size and neurons underwent tuning to identify its optimal values to be used in the experiments. Additionally, the root-mean-squared error (RMSE) is used to evaluate the performance of the model. Using the same set of training and testing data, experiment exhibits that the enhanced model achieved the optimal result of 34.805 in terms of RMSE. This means that the enhanced model outperforms the single and multiple LSTM layer with 43.5 percent and 44.95 percent reduction in error rates, respectively. Since there is no proof that LSTM recurrent neutral network has been used with the same agricultural problem domain, therefore, there is no standard available with regards to the level of error reduction in the forecast. Moreover, investigating the performance of the model using diverse datasets specifically with multiple input features (multivariate) is suggested for exploration. Furthermore, extending and embedding this approach to a web-based along with a handy application is the future plan for the benefit of the medium scale banana growers of the region for efficient and effective decision making and advance planning.

Highlights

Deep learning (DL) is a method that has been enticing attention in recent years of machine learning and its continuous growth gains more popular among researchers in diverse disciplines [1] where advancement and progression are fast and incremental
Studies have shown that agricultural problems like forecasting yields remain difficult due to the lack of the necessary infrastructures and there is no proof of optimal model to handle time series (TS) data to be used in forecasting such as the banana harvest yields dataset
There are three sets of experiment done in forecasting using banana harvest yield dataset: first, using single long short term memory (LSTM) layer, second, using multiple LSTM layers assigned with same value of neurons in each layer and third, using the enhanced deep learning model where multiple layers feed with multiple value of neurons

Summary

INTRODUCTION

Deep learning (DL) is a method that has been enticing attention in recent years of machine learning and its continuous growth gains more popular among researchers in diverse disciplines [1] where advancement and progression are fast and incremental. More accurate forecasts of the harvest yields and crop production provides an aid to an effective and efficient decision making using timely information It is a significant phase for an emerging economy so that adequate planning is undertaken for sustainable growth [4] and for the overall development of the country. Its shallow architecture makes it incapable to exemplify the complex characteristics of TS data in handling extremely nonlinear and long interval TS datasets [15] such as in banana harvest yields data This limitation compels LSTM to be unclear if it is the best design to work out real problem especially in using the harvest yield dataset and the optimization issues due to the size of the data and the model tuning strategy applied.

Deep Learning Approach

THE ENHANCED APPROACH

Data Preprocessing

Training and Forecasting

EXPERIMENTAL RESULTS AND DISCUSSION

CONCLUSION