A CONCEPTUAL MODEL FOR SEARCHING RELEVANT SCIENTIFIC PUBLICATIONS IN LARGE DATABASES USING AGENT-BASED TEXT MINING

The size of online content keeps growing dramatically with the development of scientific and technological innovations and the spread of knowledge. Searching for relevant publications has not been easy. For some submitted queries, search engines may return hundreds of documents of questionable relevance. In this paper, we propose a conceptual model to search relevant scientific publications in large databases using agent-based text mining. The proposed model provides a new search platform, whichuses text mining and multi-agent approaches. The model consists of an interface in a multi-agent layer to capture a user input and operate and control the identification process, and a text mining layer to filter results according to the user input. We use text mining to improve the accuracy of identifying highly relevant content and the multi-agent approach to speed up the process by sharing knowledge and experience via collaboration and coordination. The proposed model would be useful in providing an alternative means of searching highly relevant content from large databases. Keywords : Text Mining, Agent-based Model, Relevant Publication