The development of science and the spread of knowledge coincide with growing number of publications, and the volume of online content continue to grow at a rapid rate. For some submitted queries, the search engines may return thousands of documents of questionable relevancy. In this paper, we analyze the literature and identify the text mining factors that influence the identification of relevant studies. Five factors are identified which are Text Typography; Paragraph length; Term Frequency factor; Coordination; and Strict search. Subsequently, we propose an agent based-text mining model that facilitate the identification of relevant studies in big databases. The model consists of four components which are, interface, search process, parsing process, and storage. The interface provides a communication mean between a user and his/her counterpart agent (Personal Agent). In addition, it provides an input tool for user’s search preferences. The second component is the search process that is operated by a pattern matching. The third process is the parsing that is operated by a text mining algorithm. The last part is the storage that is managed by Monitor Agent. The proposed framework would be useful in providing an alternative means of searching highly relevant studies from large databases.
|Number of pages||12|
|Journal||Journal of Theoretical and Applied Information Technology|
|Publication status||Published - 30 Jun 2018|
All Science Journal Classification (ASJC) codes
- Theoretical Computer Science
- Computer Science(all)