“…Therefore, our emphasis in this paper is science textual data, and text mining techniques will be used to handle the raw data for feature extraction. In [20], we developed a Term Clumping process to retrieve key terms from science textual data by term removal, consolidation, and clustering. However, the aim of the original Term Clumping is to remove common terms and select core terms; thus, we undertake modifications on this approach to meet our exact needs for feature extraction, to evolve it by adding a thesaurus for our specified technical domain, adjusting the existing Science and Academic Thesaurus, skipping the Term Frequency Inverse Document Frequency (TFIDF) approach for term removal, but applying it for feature weighting, and skipping the clustering-based term consolidation approaches (e.g., Term Clustering and Combine Term Network).…”