Classification of movie review belongs to the domain of text classification, particularly in the field of sentiment analysis. Popular text classification methods for the process include Support Vector Maching (SVM) and Naïve Bayes. Both methods are known to have good performance in handling text classification individually separately. Combination of the two may be expected to improve the classification performance compared to the performance of each individual method. This paper reports an effort to classify movie review using the combined method of SVM with Naïve Bayes as the weighting factor, which is commonly called NBSVM. Our work shows that higher accuracy is obtained when classification is done using NBSVM rather than using individual methods. Accuracy at the level of 88.8% is attained when using the combined feature of unigram and bigram with only data cleansing in the pre-processing stage.
CITATION STYLE
Zain, F. F., & Sibaroni, Y. (2019). Effectiveness of SVM Method by Naïve Bayes Weighting in Movie Review Classification. Khazanah Informatika : Jurnal Ilmu Komputer Dan Informatika, 5(2), 108–114. https://doi.org/10.23917/khif.v5i2.7770
Mendeley helps you to discover research relevant for your work.