Authors: Sana'a A. Alwidian; Hani A. Bani-Salameh; Ala'a N. Alslaity
Addresses: Department of Computer Science and Applications, Faculty of Prince Al-Hussein Bin Abdallah II for IT, The Hashemite University, Zarqa 13115, Jordan ' Department of Software Engineering, Faculty of Prince Al-Hussein Bin Abdallah II for IT, The Hashemite University, Zarqa 13115, Jordan ' Department of Computer Science, Faculty of Computer and Information Technology, Jordan University of Science and Technology, Irbid 22110, Jordan
Abstract: With the increased advancements in technology and the emergence of different kinds of applications, the amount of available data becomes enormous, and the large proliferation of such data becomes evident. Therefore, there is an essential need for some techniques or methods to interact with data and extract useful information and patterns from them. Text data mining (TDM) is the process of extracting desired information out of mountains of textual data that are inherently unstructured, without the need to read them all. In this paper, we shed the light on the-state-of-the-art in text mining as an interdisciplinary field of several related areas. To facilitate the understanding of text data mining, this paper proposes a framework that visualises this field in a step-wise manner, taking into consideration the semantic of the extracted text. In addition, this paper surveys a number of useful applications and proposes a new approach for spam detection based on the proposed TDM framework.
Keywords: text mining; clustering; categorisation; spam filtering; semantics; information retrieval; text data mining; TDM; natural language processing; NLP; knowledge discovery from databases; KDD; knowledge discovery from text; KDT; semantic analysis.
International Journal of Business Information Systems, 2015 Vol.18 No.2, pp.127 - 140
Published online: 28 Mar 2015 *Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article