SUHARTO, Edy and Widodo, Aris Puji and Suryono, Suryono (2018) ANALISIS AKURASI PENGISIAN LEMBAR JAWAB UJIAN BERBASIS KERTAS MENGGUNAKAN ALGORITMA DECISION TREE C4.5. Masters thesis, School of Postgraduate.
| PDF 668Kb | |
| PDF 120Kb | |
| PDF 448Kb | |
| PDF 597Kb | |
PDF Restricted to Repository staff only 3522Kb | ||
| PDF 105Kb | |
| PDF 108Kb | |
PDF Restricted to Repository staff only 312Kb |
Abstract
Dalam penelitian ini telah dilakukan analisis terhadap data hasil pemrosesan lembar jawab ujian berbasis kertas menggunakan teknik data mining. Teknik tersebut dapat digunakan untuk melakukan klasifikasi data dan untuk menemukan komponen data penting dalam lembar jawab. Penelitian ini didasarkan pada adanya persoalan kesalahan pengisian data dalam model ujian berbasis kertas yang tidak muncul pada ujian berbasis komputer, sementara akurasi pengisian data ujian penting untuk penjaminan mutu pendidikan. Pada penelitian ini diusulkan metode untuk menganalisis akurasi pengisian data ujian di lembar jawab, yaitu menggunakan Algoritma Decision Tree C4.5. Hasil penelitian ini adalah sebuah program berbasis web untuk sistem pra-pemrosesan data dan sebuah model pohon yang mencerminkan kondisi data. Analisis terhadap 374 instans menghasilkan akurasi pengisian lembar jawab sebesar 95,19% dan akurasi klasifikasi sebesar 100%. Hasil penelitian diharapkan dapat memotivasi penyelenggara ujian untuk perbaikan sistem ujian. Kata kunci : data mining, klasifikasi, algoritma C4.5, dan ujian berbasis kertas In this study was conducted analysis on the result of a paper-based test using data mining technique. This technique is applicable to classify data and to learn significant components of answer sheet data. This study was motivated by the problem of incorrect data content in paper-based test nonexistent in computerbased test. Meanwhile, accuracy in filling out the test data is important for quality assurance in education. In this study was proposed a method to analyze the accuracy of answer sheet filling out. To do so, C4.5 Decision Tree Algorithm was used here. The results produced in this study were a web-based program for data preprocessing and a decision tree model representing data behaviour. Based on analysis of 374 instances, there were found that the accuracy of answer sheet filling out reached 95.19% and the accuracy of data classification reached 100%. The results was meant to motivate the test administrator for test improvement. Keywords : data mining, classification, C4.5 algorithm, and paper-based test
Item Type: | Thesis (Masters) |
---|---|
Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science |
Divisions: | School of Postgraduate > Master Program in Information System |
ID Code: | 66143 |
Deposited By: | INVALID USER |
Deposited On: | 26 Oct 2018 10:00 |
Last Modified: | 26 Oct 2018 10:00 |
Repository Staff Only: item control page