<![CDATA[Indonesia is a country that has many languages, in addition to Indonesian which is used as a language of communication, every region in Indonesia also has its own regional language. The number of languages owned makes it difficult for outsiders or foreigners to identify the origin of the language used, the purpose of this research is to identify languages using the nave Bayes method and random forest from the results of language identification according to the text of the Toraja, Kalimantan and Halmahera languages using computer technology. marchine learning to calculate the accuracy value of the two methods to compare the most effective methods to identify language. The results of the Naïve Bayes method in identifying language are very good because they get an accuracy value above 0.90 compared to Random Forest only getting an accuracy value below 0.70. By calculating the confusion matrix, the Naïve Bayes method is more effective with an accuracy value of 0.9922 compared to an accuracy random Forest value of 0.6544.]]>
Copyrights © 2022
