MANGA RECOMMENDATION USING NAIVE BAYES AND DECISION TREE ALGORITHM

RAHARJO, DANIEL ADRIAN (2022) MANGA RECOMMENDATION USING NAIVE BAYES AND DECISION TREE ALGORITHM. Other thesis, Universitas Katholik Soegijapranata Semarang.

[img] Text
19.K1.0054-DANIEL ADRIAN RAHARJO_COVER_a.pdf

Download (965kB)
[img] Text
19.K1.0054-DANIEL ADRIAN RAHARJO_BAB I_a.pdf

Download (189kB)
[img] Text
19.K1.0054-DANIEL ADRIAN RAHARJO_BAB II_a.pdf
Restricted to Registered users only

Download (195kB)
[img] Text
19.K1.0054-DANIEL ADRIAN RAHARJO_BAB III_a.pdf

Download (189kB)
[img] Text
19.K1.0054-DANIEL ADRIAN RAHARJO_BAB IV_a.pdf

Download (258kB)
[img] Text
19.K1.0054-DANIEL ADRIAN RAHARJO_BAB V_a.pdf

Download (728kB)
[img] Text
19.K1.0054-DANIEL ADRIAN RAHARJO_DAPUS_a.pdf

Download (195kB)
[img] Text
19.K1.0054-DANIEL ADRIAN RAHARJO_LAMP_a.pdf

Download (368kB)

Abstract

Manga is a part of Japanese culture that is very popular with people. In reading manga, of course everyone has their own criteria. Therefore, a good recommendation system is needed to assist users in finding the manga that suits their preferences. To create a recommendation system, the algorithm commonly used is Naive Bayes. However, can we use J48 to create a recommendation system? With manga data in myanimelist, this project will compare which algorithm gives the best results when used in a recommendation system. J48 can be used to make a recommendation system, but the recommendations generated from j48 are very few compared to naive bayes. The results of Naive Bayes' recommendations are more varied than J48 even though J48 provides recommendations where the data shows that the manga is indeed a good manga. When testing on the action genre with a total of 162 manga, naive bayes gave 100 recommendation results while the decision tree was 78.

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 04:12
Last Modified: 23 Mar 2022 04:12
URI: http://repository.unika.ac.id/id/eprint/28281

Actions (login required)

View Item View Item