DERIVATION OF PERTINENT ABILITIES FOR PILOTS USING LSA & SVD: A TEXT MINING APPROACH

This paper discusses an experimental study on ‘abilities required for the pilots’ data using the Latent Semantic Analysis (LSA)/ Singular Value Decomposition (SVD) technique for text extraction. LSA has been used for structuring the documents as Term Document Matrix (TDM). Term-Frequency-Inverse Document Frequency (Tf-idf) has been used for weighting the terms in TDM. Then Singular Value Decomposition (SVD) has been applied for dimension reduction and calculated U dot S to get the importance of the terms in the whole documents corpus. Finally, words are grouped using cosine similarity method. 14 abilities have been derived as a result of the study.