How to undertake reviews of large collections of articles and establish main contributions: an ontology-based literature review approach

Abstract

Identifying the main contributions of scientific works may demand time and previous depth knowledge, besides it can be a bottleneck for reviewing processes. To make it easier, this paper presents a standard representation based on the Knowledge Discovery in Databases process to characterize knowledge-discovery-in-databases scientific documents. This first attempt consists of a string composed of five elements, which provide: (i) the complexity of the problem; the tasks from the (ii) preprocessing, (iii) processing, and (iv) post-processing step, and (v) the auxiliary approaches such document used to achieve its objectives. As all documents can be classified using this approach, this proposal is scalable by text mining algorithms, making the review working easier for scientists since this methodology saves the time of the document's main classification.

Keywords

• Standard representation
• Knowledge representation
• Ontology
• Knowledge discovery in databases
• Data mining
• Pattern recognition