Perbandingan Algoritma Klasifikasi Naïve Bayes, C4.5 Dan Knn Untuk Menentukan Perokok Aktif Dan Perokok Pasif

Muhammad, Isra Muntaha Tanjung (2023) Perbandingan Algoritma Klasifikasi Naïve Bayes, C4.5 Dan Knn Untuk Menentukan Perokok Aktif Dan Perokok Pasif. Undergraduate Thesis thesis, Institut Teknologi Telkom Purwokerto.

[img] Text
Cover (1).pdf

Download (1MB)
[img] Text
Abstract (1).pdf

Download (13kB)
[img] Text
Abstrak (1).pdf

Download (12kB)
[img] Text
BAB I (1).pdf

Download (112kB)
[img] Text
BAB II (1).pdf

Download (272kB)
[img] Text
BAB III (1).pdf

Download (269kB)
[img] Text
BAB IV (1).pdf
Restricted to Registered users only

Download (630kB) | Request a copy
[img] Text
BAB V (2).pdf

Download (31kB)
[img] Text
Daftar Pustaka.pdf

Download (104kB)


Smoking is a common habit in many countries, apart from developed countries, it has also become a habit in developing countries, especially in Indonesia. According to the smoke inhaled, smokers can be divided into active smokers and passive smokers. Active smokers can be classified based on the number of cigarettessmoked per day. Passive smoking is defined as someone who is exposed to cigarette smoke for more than 15 minutes a day for more than 1 day a week. WHO (World Health Organization), estimates that in 2025 the number of smokers in Indonesia will increase by around 45% of the total population. Data mining is able to help classify whether a person belongs to the category of passive or active smokers, with symptoms in smokers or indications for a smoker. The process of analyzing active and passive smokers is carried out by a classification process and the result isthat the person is an active or passive smoker. This study uses 3 data mining algorithms namely Decision Tree, K-Nearest Neighbor and Naive Bayes. The dataset used is the Body Signal of Smoking from Kaggle. From the results of the study successfully implemented the Decision Tree algorithm (C4.5), K-Nearest Neighbor and Naïve Bayes using the Body Signal of smoking (Kaggle) dataset in predicting active and passive smoking. From the results of this study the accuracy produced by the C4.5 algorithm is 70.78%, the K-NN algorithm is 71.76% and the Naïve Bayes algorithm is 71.19%. From a comparison of the three algorithms it was found that the K-NN algorithm is the algorithm with the highest level of accuracy so that the K-NN algorithm is suitable for use in the classification of determining active and passive smokers. Keywords: Smokers, Decision Tree, Naive Bayes, K-Nearest Neighbor.

Item Type: Thesis (Undergraduate Thesis)
Subjects: T Technology > TA Engineering (General). Civil engineering (General)
Divisions: Faculty of Informatics > Informatics Engineering
Depositing User: pustakawan ittp
Date Deposited: 30 Mar 2023 02:49
Last Modified: 30 Mar 2023 02:49

Actions (login required)

View Item View Item