Expert-Based Text Mining with Fuzzy Delphi Method for Crude Oil Price Prediction
S. Chuaykoblap, P. Chutima, A. Chandrachai, N. Nupairoj
Available Online November 2015.
- https://doi.org/10.2991/itms-15.2015.92How to use a DOI?
- Knowledge Management; Forecasting; Data Mining
- As crude oil is becoming even more important commodity in the world economy, the crude oil prices fluctuation has affected both many businesses’ decision based on the crude oil prices and consumers by price inflation on consumer goods. Accordingly, the crude oil prices forecasting has continuously been interested by a lot of researchers. Text data mining from news articles is one of the widely used methods to predict the crude oil price variation caused by irregular events. Nevertheless, the main issues in text mining originate from the particularities of natural language and a great number of noisy data preventing some techniques from employing efficiently . In this research, the expert-based text mining was proposed in order to screen out the noisy data before the following steps of text mining started and to cope with the concern of natural language in text mining methods. Next, the fuzzy Delphi method was then combined in order to achieve fuzzy weighted ratings for different corresponding factors extracted from the news articles. As a result, Expert-Based Text Mining with Fuzzy Delphi Method was shown to outperform the original in term the ability to predict the historical crude oil price data.
- Open Access
- This is an open access article distributed under the CC BY-NC license.
Cite this article
TY - CONF AU - S. Chuaykoblap AU - P. Chutima AU - A. Chandrachai AU - N. Nupairoj PY - 2015/11 DA - 2015/11 TI - Expert-Based Text Mining with Fuzzy Delphi Method for Crude Oil Price Prediction BT - 2015 International Conference on Industrial Technology and Management Science PB - Atlantis Press SP - 387 EP - 391 SN - 2352-538X UR - https://doi.org/10.2991/itms-15.2015.92 DO - https://doi.org/10.2991/itms-15.2015.92 ID - Chuaykoblap2015/11 ER -