TY  - THES
N1  - Muhammad Didik Rohmad Wahyudi, S.T., MT.
ID  - digilib39252
UR  - https://digilib.uin-suka.ac.id/id/eprint/39252/
A1  - Sekar Minati, NIM. 16650023
Y1  - 2020/02/13/
N2  - Al-Qur'an is a way of life for Muslims which is a source of knowledge, policy, and law. In order to make it easier for the public to learn and understand the Al-Qur'an, one method of interpretation is namely thematic, and the Al-Qur'an index was created to make it easier to search for verses according to its theme.
This research to improve the learning outcomes of Qur?an's Ministry of Religion Republic Indonesia and Al-Qur'an Al-Hadi thematic Qur?an classification using ensemble learning methods: Bootstrap Aggregating (Bagging) and Adaptive Boosting (AdaBoost). Both of these methods tested by doing prediction to searching the proof related to Islamic articles.
Bagging and AdaBoost methods can improve classification learning outcomes with the base estimator algorithm Decision Tree classification with the average test results of the two indices obtained the value of precision 0.49, recall 0.48, f1-score 0.49, accuracy 54%, and cross-validation evaluation of 31% in Bagging method and precision 0.25, recall 0.18, f1-score of 0.16, accuracy 34%, and evaluation of cross-validation of 32% on AdaBoost.
PB  - UNIVERSITAS ISLAM NEGERI SUNAN KALIJAGA YOGYAKARTA
KW  - text mining
KW  -  classification
KW  -  ensemble
KW  -  bagging
KW  -  adaboost
KW  -  al-quran
KW  -  thematic
KW  -  Islamic articles
M1  - skripsi
TI  - STUDI KOMPARASI ENSEMBLE METHOD: BOOTSTRAP AGGREGATING (BAGGING) DAN ADAPTIVE
AV  - restricted
EP  - 133
ER  -