Recommendation for Classification of News Categories Using Support Vector Machine Algorithm with SVD

Abstract

Online news is a digital information media currently has a very easy and flexible updating process. The News Document grouping process is implemented in several stages, including Text Mining which includes Text Pre-processing which includes Tokenizing, Stopword removal, Stemming, Word Merging, TF-IDF and Confusion Matrix. Of the several techniques in Text Mining, the most frequently used for News Document classification is the Support Vector Machine (SVM). SVM has the advantage of being able to identify separate hyperplane that maximizes the margin between two or more different classes. The selection of features in SVM significantly affects the classification accuracy results. Therefore, in this study a combination of feature selection methods is used, namely Singular Value Decomposition in order to increase accuracy and reduce the Classifier Time Support Vector Machine. This research resulted in text classification in the form of categories Entertainment, Health, Politics and Technology. Based on the Support Vector Machines Algorithm, an accuracy rate of 81% was obtained with 360 Data Training and 120 Data Testing, after adding the Singular Value Decomposition feature with a K- Rank value of 50%, a significant increase in accuracy was obtained with an accuracy value of 94% and The time of Algorithm process is faster.

An Informetric View to the Negative Capacitance Phenomenon at Interlayered Metal-Semiconductor Structures and Distinct Electronic Devices

“…It is necessary to say when trained SVD and then K-means are used together or just K-means and elbow method are used in the algorithm they give closer numbers to WOS results but since the labels should be clear up to the sub-categories, it was not preferred. Since a study showed that a combination of SVD and SVM algorithms gave better accuracy when used together than just SVM used itself, we preferred a similar way as in their work by employing the favors of SVD by dimension reduction (., & Rarasati , 2022). When using the data tables for both training and testing, we employed the same index terms for training as the same as in the WOS search.…”