Job Seeker Profile Classification of Twitter Data Using the Naïve Bayes Classifier Algorithm Based on the DISC Method

Human resource staff in a company is a person in charge of finding new workers. To get a qualified new workforce, a human resource staff must be selective toward the appliance in terms of ability and personality. This study provides an alternative perspective for a human resource in getting one's personality data through their tweets on a Twitter account. This study uses the Naive Bayes Classifier algorithm with W-IDF (Weighted-Inverse Document Frequency) weighting to classify the personality of recruits into one of DISC's personality theories, namely Dominance, Influence, Steadiness, and Compliance. By using training data and test data as many as 120 personal Twitter accounts and labelling of words that have been verified by psychologists, obtained personality distribution. The classification of the tweet data is Dominance 90 accounts, Influence 10 accounts, Steadiness 8 accounts and Compliance 12 accounts. Evaluation of the accuracy level of 36.67%.