PERBANDINGAN KINERJA METODE MACHINE LEARNING ANTARA MODEL NAIVE BAYES DAN K-NEAREST NEIGHBOR TERHADAP PERSEPSI PENGGUNA JASA TRANSPORTASI ONLINE

ADZIKRI, MUHAMAD DAFA (2022) PERBANDINGAN KINERJA METODE MACHINE LEARNING ANTARA MODEL NAIVE BAYES DAN K-NEAREST NEIGHBOR TERHADAP PERSEPSI PENGGUNA JASA TRANSPORTASI ONLINE. S1 thesis, Universitas Mercu Buana Jakarta.

Preview	Text (ABSTRAK) 02 ABSTRAK.pdf Download (61kB) | Preview
Preview	Text (HAL COVER) 01 COVER.pdf Download (1MB) | Preview
	Text (BAB I) 03 BAB 1.pdf Restricted to Registered users only Download (103kB)
	Text (BAB II) 04 BAB 2.pdf Restricted to Registered users only Download (94kB)
	Text (BAB III) 05 BAB 3.pdf Restricted to Registered users only Download (130kB)
	Text (BAB IV) 06 BAB 4.pdf Restricted to Registered users only Download (104kB)
	Text (BAB V) 07 BAB 5.pdf Restricted to Registered users only Download (115kB)
	Text (BAB VI) 08 BAB 6.pdf Restricted to Registered users only Download (303kB)
	Text (DAFTAR PUSTAKA) 09 DAFTAR PUSTAKA.pdf Restricted to Registered users only Download (82kB)
	Text (LAMPIRAN) 10 LAMPIRAN.pdf Restricted to Registered users only Download (268kB)

Abstract

Abstract − Online transportation is one of the means that makes it easier for humans to carry out daily activities. However, in the conditions of the COVID-19 pandemic, many people are afraid to use online transportation facilities, which causes a decrease in public interest in online transportation. After the covid-19 pandemic was over, entering an endemic period, this thing made researchers interested in seeing how the community responded through tweets on the twitter application. In this study, the process of collecting data with keywords relevant to online transportation was carried out. Then the labeling process is carried out manually and automatically in order to find out the best way to label the data. Based on the dataset, a predictive model was developed using the naive bayes method and k-nearest neighbor as a classification method. The method was tested with three experiments with the distribution of data as much as 10%, 20% and 30% in order to find out the best distribution of the dataset in achieving the accuracy value obtained. Based on the experiments that have been carried out, the nave Bayes method got the highest accuracy as much as 95% using manual labels on data separation as much as 10%, and the k-nearest neighbor method got the highest accuracy as much as 93% using manual labels on data separation as much as 10%. Key words: Transportasi Online, Naïve Bayes, K-Nearest Neighbor, Machine Learning Abstrak − Transportasi online merupakan salah satu sarana yang memudahkan manusia dalam melakukan aktivitas sehari-hari. Namun, dalam kondisi pandemi covid-19 banyak sekali masyarakat yang takut untuk menggunakan sarana transportasi online yang menyebabkan penurunan minat masyarakat pada transportasi online. Setelah pandemi covid-19 usai, memasuki masa endemi hal ini yang membuat peneliti tertarik untuk melihat bagaimana respon masyarakat melalui cuitan pada aplikasi twitter. Pada penelitian ini dilakukan proses pengumpulan data dengan kata kunci yang relevan dengan transportasi online. Kemudian dilakukan proses label secara manual dan secara otomatis guna mengetahui cara terbaik untuk melabel data. Berdasarkan dataset tersebut, dikembangkan model prediktif dengan menggunakan metode naive bayes dan knearest neighbor sebagai metode klasifikasi. Metode di uji coba dengan tiga percobaan dengan pembagian data sebanyak 10%, 20% dan 30% guna mengetahui pembagian dataset terbaik dalam mencapai nilai akurasi yang didapat. Berdasarkan percobaan yang telah dilakukan metode naïve bayes mendapat akurasi tertinggi sebanyak 95% dengan menggunakan label manual pada pemisahan data sebanyak 10%, dan metode k-nearest neighbor mendapat akurasi tertinggi sebanyak 93% menggunakan label manual pada pemisahan data sebanyak 10%. Kata Kunci : Transportasi Online, Naïve Bayes, K-Nearest Neighbor, Machine Learning.

Actions (login required)

View Item