Komparasi akurasi naïve bayes dan support vector machine (svm) untuk rekomendasi produk in fashion dress
Total View This Week0
Total View This Week0
Institusion
Institut Teknologi Telkom Purwokerto
Author
Nancy, Ria Silvani Hutauruk
Subject
T Technology (General)
Datestamp
2021-04-26 03:48:39
Abstract :
Data mining or data discovery is a method that uses statistical techniques and data analysis using data that has meaning from a broad and interesting data set to useful information. In data mining there are several methods in data mining, one of which is classification. Classification becomes one of the data mining techniques used to model data from sample data that has not been classified to classify new sample data into similar classes. In previous studies, it was found that SVM was better than Naïve Bayes in classifying, so the researchers compared SVM and NB algorithms with different case studies, namely products in clothing. Naïve Bayes because it has advantages is a simple algorithm but has high accuracy. And SVM also has the advantage that svm also has a high level of accuracy and can work very well on data with various dimensions and dimensions of dimensions. The purpose of this research is to apply and analyze the results using the Naïve Bayes algorithm and SVM in making clothes products. The results obtained from this study are Naive Bayes better than SVM, where the results on Naive Bayes are 74% and in SVM is 66%. The Matrix tool for the classification process in the Dresses_Attribute_Sales dataset is given via the UCI Repository. Keywords - Dataset , Naive Bayes, SVM, Classification, Dress
Data mining or data discovery is a method that uses statistical techniques and data analysis using data that has meaning from a broad and interesting data set to useful information. In data mining there are several methods in data mining, one of which is classification. Classification becomes one of the data mining techniques used to model data from sample data that has not been classified to classify new sample data into similar classes. In previous studies, it was found that SVM was better than Naïve Bayes in classifying, so the researchers compared SVM and NB algorithms with different case studies, namely products in clothing. Naïve Bayes because it has advantages is a simple algorithm but has high accuracy. And SVM also has the advantage that svm also has a high level of accuracy and can work very well on data with various dimensions and dimensions of dimensions. The purpose of this research is to apply and analyze the results using the Naïve Bayes algorithm and SVM in making clothes products. The results obtained from this study are Naive Bayes better than SVM, where the results on Naive Bayes are 74% and in SVM is 66%. The Matrix tool for the classification process in the Dresses_Attribute_Sales dataset is given via the UCI Repository. Keywords - Dataset , Naive Bayes, SVM, Classification, Dress
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
