Please use this identifier to cite or link to this item:
|Title:||Hybrid ARIMAX and LSTM Model to Predict Rice Export Price in Thailand|
|Keywords:||Computer Science;Decision Sciences|
|Abstract:||Rice is an important export product of Thailand. In addition, Thailand is one of the top 3 world rice exporters. This research proposes a hybrid model to predict the export price of Hom Mali Rice and White Rice. The proposed model includes three processes. Firstly, input features are prepared with Extract-Transform-Load. Secondly, a Genetic algorithm is used to select only important input features. Finally, Autoregressive Integrated Moving Average with Explanatory Variable (ARIMAX) is created as the new input feature by transforming original input features, and then prediction models are created from Long Short-Term Memory (LSTM). The results of the proposed model to predict data between 2016 and 2020 showed that this model has an average of 10.5531 and 9.3132 of Root Mean Square Error and 7.9152 and 7.6999 of the Mean Absolute Error for Hom Mali Rice and White Rice, respectively. Moreover, the proposed model outperforms when compared with six other prediction models.|
|Appears in Collections:||CMUL: Journal Articles|
Files in This Item:
There are no files associated with this item.
Items in CMUIR are protected by copyright, with all rights reserved, unless otherwise indicated.