PENGOLAHAN TEKS NARATIF BAKU MENGGUNAKAN LSA DAN LEXICAL TAXONOMY SEBAGAI INPUT SISTEM PAKAR PENYAKIT THT; FORMAL NARRATIVE TEXT PROCESSING UTILIZING LSA AND LEXICAL TAXONOMY AS INPUT OF OTOLARYNGOLOGI DISEASES EXPERT SYSTEMS

RINA SEPTIRIANA, Retantyo Wardoyo

2015 | Disertasi | FMIPA

An user interface of expert systems is a means of communication an between expert systems and user. the input for user interface of expert systems doesn’t only in the form check-box or yes/no question but also images, voices, etc based on the requirement of expert systems. In this research, the problem to overcome is how to develop a user interface a processing of free texts to become a set of fact and can be treated by expert systems to produce a conclusion. The method that will be used in this study is LSA with the structure of Lexical Taxonomy. LSA is a method that is used to find a similarity between word or sentence and represent their relation into matrix form. The value of similarity is obtained from calculation of Singular Value Decomposition and the similarity cosine. Lexical Taxonomy has a role to do an indexing processing between free texts and selected word from symptom of knowledge based of expert systems. Some tests for the system show that the average value of F-score for using of LSA with Lexical Taxonomy structure is smaller than average value of F-score on LSA using only.

Kata Kunci : user interface of expert systems; latent semantic analysis; singular value decomposition; lexical taxonomy.