Overcoming Data Imbalance Problems in Sexual Harassment Classification with SMOTE

Delivery of justice with the help artificial intelligence is a current research interest. Machine learning natural language processing (NLP) can classify types sexual harassment experiences into quid pro quo (QPQ) and hostile work environments (HWE). However, imbalanced data are often present in classes classification on specific datasets. Data imbalance cause decrease classifier's performance because it usually tends to choose majority class. This study proposes implementation evaluation synthetic minority over-sampling technique (SMOTE) improve QPQ HWE classifications experience dataset. The term frequency-inverse document frequency (TF-IDF) method applies weighting process. Then, we compare naïve Bayes K-Nearest Neighbor (KNN) classifying experiences. comparison shows that classifier superior KNN HWE, AUC values 0.95 versus 0.92, respectively. results show by applying SMOTE classifier, precision class increase from 74% 90%.