CAHYANI, DENIS NILA (2024) PENERAPAN DATA MINING MENGGUNAKAN ALGORITMA NAIVE BAYES, C4.5, DAN K-NEAREST NEIGHBOR UNTUK KLASIFIKASI KEMISKINAN DI DKI JAKARTA. S1 thesis, Universitas Mercu Buana Jakarta.
![[img]](https://repository.mercubuana.ac.id/style/images/fileicons/text.png) |
Text (HAL COVER)
01 COVER.pdf Download (616kB) |
|
Text (ABSTRAK)
02 ABSTRAK.pdf Download (142kB) |
|
Text (BAB I)
03 BAB 1.pdf Restricted to Registered users only Download (214kB) |
|
Text (BAB II)
04 BAB 2.pdf Restricted to Registered users only Download (279kB) |
|
Text (BAB III)
05 BAB 3.pdf Restricted to Registered users only Download (303kB) |
|
Text (BAB IV)
06 BAB 4.pdf Restricted to Registered users only Download (1MB) |
|
Text (BAB V)
07 BAB 5.pdf Restricted to Registered users only Download (146kB) |
|
Text (DAFTAR PUSTAKA)
08 DAFTAR PUSTAKA.pdf Restricted to Registered users only Download (205kB) |
|
Text (LAMPIRAN)
09 LAMPIRAN.pdf Restricted to Registered users only Download (739kB) |
Abstract
Poverty remains one of the fundamental problems that is difficult to overcome, including in DKI Jakarta Province as the capital of Indonesia, which is also not immune to poverty. This research uses data mining to classify the poverty level in DKI Jakarta Province with data obtained from the Central Bureau of Statistics (BPS). The purpose of this study is to effectively and accurately classify poverty data and compare the results of the Naive Bayes, KNN, and C4.5 algorithms to find the best accuracy results. Through the application of such algorithms, the results were obtained that the performance of the algorithm could be influenced by the division of data, and increased ratio of test data tended to improve the accuracy as well as consistency of classification results. At the data division ratio of 70:30, Naïve Bayes achieved accuracy of 81%, C4.5 of 76%, and K-NN of 71%. At the ratio of 80:20, Naïve Bayes showed accurateness of 93%, C4.5 of 79%, and K-NN of 86%. Whereas at the ratios of 90:10, Naïve Bayes achieves accurateness of 100%, C4.5 of 71%, and K-NN of 86%. By considering the performance variations of the third algorithm, it can be concluded that Naïve Bayes is superior as a stable and reliable algorithm in a variety of data set splitting scenarios, showing high accuracy and good ability in identifying positive cases. Keywords: poverty, classification, Naïve Bayes, C4.5, K-Nearest Neighbor Kemiskinan masih menjadi salah satu permasalahan fundamental yang sulit dihadapi, termasuk di Provinsi DKI Jakarta sebagai Ibu kota negara Indonesia, yang juga tidak luput dari permasalahan kemiskinan. Penelitian ini menggunakan data mining untuk melakukan klasifikasi terhadap tingkat kemiskinan di Provinsi DKI Jakarta dengan data yang diperoleh dari Badan Pusat Statistik (BPS). Tujuan penelitian ini adalah untuk secara efektif dan akurat mengklasifikasikan data kemiskinan dan membandingkan hasil dari algoritma Naïve Bayes, C4.5N dan K-NN guna mencari hasil akurasi terbaik. Melalui penerapan algoritma tersebut, didapatkan hasil bahwa performa algoritma dapat dipengaruhi oleh pembagian data, dan peningkatan rasio data uji cenderung meningkatkan akurasi serta konsistensi hasil klasifikasi. Pada rasio pembagian data 70:30, Naïve Bayes mencapai akurasi 81%, C4.5 sebesar 76%, dan K-NN sebesar 71%. Pada rasio 80:20, Naïve Bayes menunjukkan akurasi 93%, C4.5 sebesar 79%, dan K-NN sebesar 86%. Sementara pada rasio 90:10, Naïve Bayes mencapai akurasi 100%, C4.5 sebesar 71%, dan K-NN sebesar 86% Dengan mempertimbangkan variasi performa ketiga algoritma, dapat disimpulkan bahwa Naïve Bayes lebih unggul sebagai algoritma yang stabil dan dapat diandalkan dalam berbagai skenario pembagian dataset, menunjukkan akurasi tinggi dan kemampuan baik dalam mengidentifikasi kasus positif. Kata kunci: kemiskinan, klasifikasi, Naïve Bayes, C4.5, K-Nearest Neighbor
Actions (login required)
 |
View Item |
