PENERAPAN MEMBERSHIP DEGREE TRANSFORMATION NEW ALGORITHM M(1,2,3) UNTUK MENGEVALUASI KEPUASAN PELAYANAN MAHASISWA PASCASARJANA (STUDI KASUS : PASCASARJANA ILMU KOMPUTER UGM)

RIAH UKUR GINTING, Drs. Retantyo Wardoyo, M.Sc., Ph.D.

2011 | Tesis | S2 Ilmu Komputer

Banyak aspek yang perlu dipertimbangkan saat melakukan proses evaluasi kepuasan pelayanan suatu institusi. Masalah yang sering muncul adalah terjadinya duplikasi data dan ambiguitas sehingga perlu diterapkan sebuah metode evaluasi fuzzy secara komprehensif berbasis membership degree transformation. Namun pendekatan ini memiliki kelemahan yaitu turut dihitungnya duplikasi data pada indeks membership degree, padahal duplikasi data tersebut tidak berguna untuk hasil perhitungan. New algorithm M(1,2,3) adalah: (i) penggunaan teknologi data mining berdasarkan entropi untuk menggali informasi (knowledge) tentang klasifikasi objek yang tersembunyi dalam setiap indeks, (ii) menghilangkan duplikasi data yang terdapat dalam keanggotaan indeks dengan cara memberi bobot pembeda. New algorithm M(1,2,3) meliputi: distinguishable weight, effective value dan comparable value. Pada penelitian ini, kami mengimplementasikan new algorithm M(1,2,3) untuk mengevaluasi kepuasan pelayanan mahasiswa Pascasarjana Ilmu Komputer UGM.

In the evaluation process of institution, there are many aspects needing to consider, with a lot of uncertainty and ambiguity, so it is reasonable and scientific to apply fuzzy comprehensive evaluation method for UGM postgraduate student satisfaction evaluation. The core of fuzzy evaluation is membership degree transformation. But the existing transformation methods should be questioned, because redundant data in index membership degree is also used to compute object membership degree, which is not useful for object classification. The new algorithm is: using data mining technology based on entropy to mine knowledge information about object classification hidden in every index, affirm the relationship of object classification and index membership, eliminate the redundant data in index membership for object classification by defining distinguishable weight and extract valid values to compute object membership. The new algorithm of membership degree transformation includes three calculation steps which can be summarized as “effective, comparison and composition”, which is denoted as M(1,2,3). The paper applied the new algorithm in the fuzzy evaluation of UGM postgraduate student satisfaction.

Kata Kunci : membership degree transformation new algorithm M(1,2,3), duplikasi data, fuzzy evaluation.