KLASIFIKASI KELAYAKAN PENERIMA KARTU JAKARTA PINTAR (KJP) DENGAN SELEKSI FITUR BACKWARD ELIMINATION MENGGUNAKAN ALGORITMA KLASIFIKASI NAÏVE BAYES

• Hoofdauteur: Jamalul Ikhsan, (Auteur)
• Formaat: Boek
• Gepubliceerd in: 2022-07-31.
• Onderwerpen: Thesis; NonPeerReviewed
• Online toegang: Link Metadata
• Notes: http://repository.upnvj.ac.id/19732/1/ABSTRAK.pdf; http://repository.upnvj.ac.id/19732/13/AWAL.pdf; http://repository.upnvj.ac.id/19732/14/BAB%20I.pdf; http://repository.upnvj.ac.id/19732/15/BAB%20II.pdf; http://repository.upnvj.ac.id/19732/16/BAB%20III.pdf; http://repository.upnvj.ac.id/19732/17/BAB%20IV.pdf; http://repository.upnvj.ac.id/19732/18/BAB%20V.pdf; http://repository.upnvj.ac.id/19732/19/DAFTAR%20PUSTAKA.pdf; http://repository.upnvj.ac.id/19732/9/DAFTAR%20RIWAYAT%20HIDUP.pdf; http://repository.upnvj.ac.id/19732/10/LAMPIRAN.pdf; http://repository.upnvj.ac.id/19732/12/HASIL%20PLAGIARISME.pdf; http://repository.upnvj.ac.id/19732/11/ARTIKEL%20KI.pdf
• Keywords: T Technology (General)

The Jakarta Smart Card or commonly referred to as KJP is a program created by the provincial government of DKI Jakarta to provide educational assistance in the DKI Jakarta area. Of course, every program has a selection system, as in the Jakarta Smart Card program, the government conducts a selection process to ensure that people who deserve assistance and who don't. In the acceptance selection process sometimes there is an omission by the operator in inputting data so that the community should be able to get help but because of this negligence they cannot get help, besides that there is a subjective selection process so that the assistance provided by the government is often not on target. Therefore, this study aims to create a system that can directly and accurately determine the eligibility of Jakarta smart card recipients using the nave Bayes classification algorithm and backward elimination feature selection to determine the features that play an important role in determining the eligibility of Jakarta smart card recipients. The dataset was obtained from the school with a total of 158 data on prospective recipients. Based on the results of this study, the accuracy obtained by the model and after calculating the accuracy using the confusion matrix on the model is obtained with an accuracy of 90.625%, but with the help of backward elimination feature selection, the accuracy can increase significantly to 96.875%.