PERSONAL IDENTIFICATION BERBASIS WEB MENGGUNAKAN MULTINOMIAL NAÏVE BAYES

Andreas Khosasi, Viny Christanti Mawardi, Meiske Yunitree Suparman

Abstract


Personality is a whole way an individual reacts and interacts with other individuals[1] In this personal identification using the basic theory of Hippocrates. The method used for classification is Multinomial Naïve Bayes. There are several stages of preprocessing such as casefolding, tokenizing with unigram and Bigram well as term frequency. The results of the classification is divided into four categories: sanguine, melancholic, plegmatis and choleric. Confusion matrix is used as evaluation  with 85 %  accuracy value of 160 training data and  40 testing data consist each category.

Full Text:

PDF

Refbacks

  • There are currently no refbacks.