Categorization of Exam Questions based on Bloom Taxonomy using Naïve Bayes and Laplace Smoothing

Autor:	Eka Rahayu Setyaningsih, Indah Listiowarni
Rok vydání:	2021
Předmět:	business.industry Feature selection Machine learning computer.software_genre Naive Bayes classifier Statistical classification Categorization Taxonomy (general) Bloom's taxonomy Artificial intelligence Additive smoothing business computer Test data
Zdroj:	2021 3rd East Indonesia Conference on Computer and Information Technology (EIConCIT).
DOI:	10.1109/eiconcit50028.2021.9431862
Popis:	Being famous for a classification algorithm using a simple statistic calculation, Naive Bayes produces a relatively low accuracy. This research tests how combining the Naive Bayes classifier using Chi-Square as its feature selection, accompanied by Laplace Smoothing, may improve its accuracy. The tests classify 600 high school biology exam questions in Bahasa Indonesia into Bloom’s Taxonomy of cognitive domain. Test results show an increase of 39.6% accuracy is produced from 21.03% to 60.63% when 100 new data introduced. On the other hand, a 22.19% increase is gained from 53.75% to 75.94% when a mix of testing data (10, 20, 50, 100, 250) is applied.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::a12851f536eab8aa43fded208bfdf04d https://doi.org/10.1109/eiconcit50028.2021.9431862 Zobrazit plný text záznamu