WICAKSONO, NIKODEMUS GALIH CANDRA (2022) COMPARISON LEARNING VECTOR QUANTIZATION AND NAIVE BAYES ALGORITHM IN AIRLINE PASSENGER SATISFACTION. Other thesis, Universitas Katholik Soegijapranata Semarang.
Text
18.K1.0082NIKODEMUS GALIH CANDRA WICAKSONO_COVER_a.pdf Download (406kB) 

Text
18.K1.0082NIKODEMUS GALIH CANDRA WICAKSONO_BAB I_a.pdf Download (117kB) 

Text
18.K1.0082NIKODEMUS GALIH CANDRA WICAKSONO_BAB II_a.pdf Restricted to Registered users only Download (125kB) 

Text
18.K1.0082NIKODEMUS GALIH CANDRA WICAKSONO_BAB III_a.pdf Download (120kB) 

Text
18.K1.0082NIKODEMUS GALIH CANDRA WICAKSONO_BAB IV_a.pdf Download (522kB) 

Text
18.K1.0082NIKODEMUS GALIH CANDRA WICAKSONO_BAB V_a.pdf Download (320kB) 

Text
18.K1.0082NIKODEMUS GALIH CANDRA WICAKSONO_BAB VI_a.pdf Download (114kB) 

Text
18.K1.0082NIKODEMUS GALIH CANDRA WICAKSONO_DAPUS_a.pdf Download (235kB) 

Text
18.K1.0082NIKODEMUS GALIH CANDRA WICAKSONO_LAMP_a.pdf Download (287kB) 
Abstract
Data is something that is very easy to obtain nowadays. Because data is easily obtained and stored digitally, the data becomes very large. For example, data from airline customer satisfaction which reaches around 130000 data. However, to process that much data, an algorithm is needed. Therefore, here I use the Naive Bayes algorithm and learning vector quantization (LVQ). Both algorithms are algorithms for classifying. This study was conducted to determine whether the two algorithms can be used in classifying airline passenger satisfaction data. I also compared the two algorithms to find out which one is better at classifying the airline satisfaction data. The level of accuracy becomes a parameter to determine a better algorithm. In completing this research, several steps were carried out. The first step is to get and process the data. In processing the data, data preprocessing is carried out, namely deleting empty data and changing the data so that the program can process it. After that, the classification process is carried out using both algorithms. Each algorithm test was carried out 5 tests with different amounts of training data, namely 90%, 75%, 50%, 25% and 10%. After the test is complete, the accuracy of all tests is calculated. The average accuracy of each test with different training data will be the accuracy value of the algorithm. From the results of the testing that has been done, there are 5 tests for naive Bayes and also 5 tests for learning vector quantization. The accuracy results obtained are the average Naive Bayes accuracy of 89.076% while the average accuracy of learning vector quantization is 79.39% for the airline's passenger satisfaction data. So it can be concluded that the two algorithms can be used to classify airline passenger satisfaction data and the Naive Bayes algorithm is better than learning vector quantization.
Item Type:  Thesis (Other) 

Subjects:  000 Computer Science, Information and General Works > 004 Data processing & computer science 
Divisions:  Faculty of Computer Science > Department of Informatics Engineering 
Depositing User:  mr AM. Pudja Adjie Sudoso 
Date Deposited:  23 Mar 2022 04:11 
Last Modified:  23 Mar 2022 04:11 
URI:  http://repository.unika.ac.id/id/eprint/28279 
Actions (login required)
View Item 