The movements in oil prices are very complex and, therefore, seem to be unpredictable. The continuous usage of statistical and econometric techniques for crude oil price prediction might demonstrate demotions to the prediction performance. Crude oil price prediction depends on heavily on uncertainty in the crude oil price fluctuation. The proposed approach uses a fuzzy rule based system embedded in fuzzy time series application to accurately extract a feature weight that can predict crude oil price prediction accurately. Another major parameter used for crude oil price prediction is news feeds. Our proposed approach extracts features weights from news feeds using a sentimental analysis based on latent dirichlet allocation topic model that can distinguish various online news topics. Both these feature weights along with the quantitative key factors are feed in to recurrent neural network.
Published In : IJCSN Journal Volume 8, Issue 3
Date of Publication : June 2019
Pages : 262-265
Tables : 01
Dona Sara Jacob :
received her B.Tech (Computer Science and Engineering) degree from University of Kerala, Trivandrum in 2017. She is currently pursuing his Masters in Computer Science & Engineering from KTU.
Sam G Benjamin :
is working as Assistant Professor in Computer Science and Engineering. His research interests focuses on image processing, data mining, and image mining.
Dr. Radhakrishnan B :
is working as the Head of CSE department. He has more than 14 years experience in teaching and has published papers on data mining and image processing. His research interests include image processing, data mining, image mining.
Crude oil price prediction, fuzzy rule based system, sentimental Analysis, Latent Dirichlet Allocation topic Model
This study proposes a new prediction model based on fuzzy rule based time series and sentimental analysis method. Here, the crude oil price fluctuation, online news headlines and other crude oil news headlines are considered as the input. In news feeds processing unit, the news headlines are tokenized and filtered then each headlines are designated with a weight. The weighted words undergo a sentimental analysis and features are extracted and grouped using LDA (Latent Dirichlet Allocation). In Parallel crude oil price and production are analysed based on a fuzzy learning method and both features are feed into a recurrent neural network. In recurrent neural network Long short Term Memory (LSTM) method is used for training.
 Xuerong Li, Wei Shang, Shouyang Wang. Text based crude oil price forecasting: A deep learning approach  Moshiri, Saeed and Foroutan, Faezeh, Forecasting Nonlinear Crude Oil Future Prices.  Nur Fazliana Rahim, Mahmod Othman, Rajalingam Sokkalingam, and Evizal Abdul Kadir. Forecasting Crude Palm Oil Prices using Fuzzy Rule-Based Time Series Method.  S. N. Abdullah, X. Zeng. Machine Learning Approach for Crude Oil Price Prediction with Artificial Neural Networks-Quantitative (ANN-Q) Model  Mayuree Sompui and Wullapa Wongsinlatam. Prediction Model for Crude Oil Price Using Artificial Neural Networks.