Classification of Reliable Content on Cancer Thai Website using CancerDic+

Main Article Content

สุภาพร เกิดกิจ องอาจ อุ่นอนันต์ พยุง มีสัจ

Abstract

- Nowadays, there are a lot of informative websites about cancer which enables users to access data easily. However, it is difficult to determine whether these websites are reliable. The objective of this research was to develop a method which could classify the credibility of a cancer website and then determine if the information was reliable. To achieve the above goal, this research applied CancerDic+ to extract words. Technical terms referring to cancer were added to a database, and then text mining was applied to classify inputted data. The values of accuracy, precision and recall derived from the word extraction by Lexto, SWATH and CancerDic+ were then compared. The results showed that text mining for word extraction by CancerDic+ yielded the best result (Accuracy = 0.844, Precision = 0.838, Recall =0.845). In conclusion, this classification method successfully gained high accuracy and can be used in other fields effectively.

Keywords

Article Details

How to Cite
เกิดกิจส., อุ่นอนันต์อ., & มีสัจพ. (2015). Classification of Reliable Content on Cancer Thai Website using CancerDic+. JOURNAL OF INFORMATION SCIENCE AND TECHNOLOGY, 5(2), 34-43. https://doi.org/10.14456/jist.2015.11
Section
Research Article: Soft Computing (Detail in Scope of Journal)