Classifying Historical Events Using Support Vector Machines (SVM) and Decision Trees

Archives must be classified and categorized according to their classification to ensure effective retrieval and analysis of past data in digital libraries. This study uses Support Vector Machines classification and Decision Trees to assist with the classification while combining their benefits. The existing classification methods are plagued by high-dimensional data, event classification uncertainty, and huge computational expenses that reduce retrieval precision. These constraints interfere with the applications of historical archives in digital repositories. To remedy these problems, we introduce a hybrid SVM-DT method whose core is based on SVM's ability to cope with high-dimensional data and the interpretability of DT. The method begins by extracting and preprocessing historical event features of interest, uses SVM to optimize feature discrimination, and uses DT to achieve accurate classification. Computational effectiveness is achieved with improved classification accuracy using this hybrid approach. SVM-DT methodology enhances the infrastructure of electronic historical records and therefore enhances classification, search, and accessibility. The SVM-DT system facilitates historians and researchers in retrieving associated historical events with more accuracy through an organized classification scheme. Experimental outcomes show drastic improvement in accuracy, precision, and recall while using the SVM-DT system when compared with the conventional classification system. The results corroborate that an SVM with a DT classification methodology is more scalable and efficient in digital libraries.