Deep Learning Based Prediction Model for Crude Palm Oil Prices Using News Sentiment Analysis with Sliding Window

K. Kanchymalay; U.R. Hashim; R. Krishnan; R.R.R. Ikram; R.R. Kuhan

doi:10.37965/jait.2024.0327

Deep Learning Based Prediction Model for Crude Palm Oil Prices Using News Sentiment Analysis with Sliding Window

Authors

K. Kanchymalay Faculty of Information and Communication Technology, Universiti Teknikal Malaysia Melaka,Melaka,Malaysia https://orcid.org/0000-0003-1424-1510
U.R. Hashim Faculty of Information and Communication Technology, Universiti Teknikal Malaysia Melaka,Melaka,Malaysia https://orcid.org/0000-0002-4105-2005
R. Krishnan Faculty of Business & Management, Universiti Teknologi MARA, Malaysia https://orcid.org/0000-0001-7752-2841
R.R.R. Ikram Faculty of Information and Communication Technology, Universiti Teknikal Malaysia Melaka,Melaka,Malaysia https://orcid.org/0000-0002-9845-5155
R.R. Kuhan Faculty of Computer Science, University of Duisburg-Essen, Schützenbahn 70, 45127 Essen Germany https://orcid.org/0009-0006-0165-3662

DOI:

https://doi.org/10.37965/jait.2024.0327

Keywords:

crude palm oil (CPO), price prediction, artificial neural network (ANN), news sentiment analysis, long short-term memory (LSTM)

Abstract

Crude palm oil (CPO) price prediction plays an important role in agricultural economic development. Various economics and agricultural related factors have been used to predict CPO prices. Nevertheless, understanding news sentiment features will also be important in CPO price prediction. This paper proposes a CPO price prediction model to help the plantation organizations in the palm oil sector to successfully anticipate CPO price fluctuations and manage the resources more effectively. The CPO price behavior is non-linear in nature, and thus prediction is very difficult. In this paper, an improved version of recurrent network, Long Short-Term Memory (LSTM) based CPO price prediction model with news sentiment is used to produce an enhanced prediction model. The findings of this study show that the LSTM based forecasting model with news headline sentiment using a six-month sliding window produced the best result in forecasting the CPO price movement compared to other sliding window sizes.