STUDI EKSPERIMENTASI DENGAN CLASSIFICATION BASED ON ASSOCIATION PADA KLASIFIKASI SANTRI BARU (STUDI KASUS: MADRASAH SALAFIYAH TIGA PP. AL-MUNAWWIR)

Yayah Siti Nurkomariah, NIM.: 16650046 (2020) STUDI EKSPERIMENTASI DENGAN CLASSIFICATION BASED ON ASSOCIATION PADA KLASIFIKASI SANTRI BARU (STUDI KASUS: MADRASAH SALAFIYAH TIGA PP. AL-MUNAWWIR). Skripsi thesis, UIN SUNAN KALIJAGA YOGYAKARTA.

Preview	Text (STUDI EKSPERIMENTASI DENGAN CLASSIFICATION BASED ON ASSOCIATION PADA KLASIFIKASI SANTRI BARU (STUDI KASUS: MADRASAH SALAFIYAH TIGA PP. AL-MUNAWWIR)) 16650046_BAB I_IVatau V_DAFTAR PUSTAKA.pdf - Published Version Download (2MB) | Preview
	Text (STUDI EKSPERIMENTASI DENGAN CLASSIFICATION BASED ON ASSOCIATION PADA KLASIFIKASI SANTRI BARU (STUDI KASUS: MADRASAH SALAFIYAH TIGA PP. AL-MUNAWWIR)) 16650046_BAB II_sampai_SEBELUM BAB TERAKHIR.pdf - Published Version Restricted to Registered users only Download (6MB) | Request a copy

Abstract

Acceptance of new students at the Salafiyah Tiga Madrasah is carried out every new school year by taking the New Santri Entrance Examination (UMSB). Data processing that is done now only uses data of the current school year not using data of the previous school year. Also, only using the entrance examination scores in the determination of new santri classes, so that there are still many santri in their learning did not develop optimally. Thus, the researcher analyzes the association between the entrance examination scores and the final exam scores as learning outcomes and uses the results of the association analysis as classification material. The purpose of this study is to know the association of the entrance examination scores for the new school year and final exam scores and to determine the class classification based on the results of the entrance examination scores. This study utilizes data mining techniques using the Classification Based on Association algorithm which applies association and classification. This algorithm has the stages of producing rule items that are often CBA-RG, applying the CBA-RG algorithm to produce CAR sets, then building a classifier based on CARs sets. This algorithm can produce a model for classifying new Santri of Madrasah Salafiyah Tiga in class determination divided into classes I'dad, Awwal, Tsani, and Tsalits. The results of testing the CBA algorithm with Python get ENOUGH rule → Awwal with the highest confidence value of 0.97 or 97%, meaning that of all students who get enough grades then get a starting class of 0.97 or 97%. Whereas the calculation of the average number of rules in the classifier built by CBA-CB uses prCAR. The first value is the average number of M1 more efficient than M2 with an average of 78. Also, testing accuracy in Python using the CBA algorithm with support = 0.01, and confidence = 0.5 produces an accuracy of 74.37379595761079%.

Share this knowledge with your friends :

Actions (login required)

View Item