Pemodelan Klasifikasi Gaji Menggunakan Support Vector Machine

Abstract

It is known that there are currently many types of work in the field. Creativity of the community and economic pressure that is felt makes people have to work hard to be able to meet the needs of life. One way that must be done to be able to continue to survive by working. By working someone can produce wages or salaries so that the necessities of life of a person can be met. Various work that exists raises a problem. In determining the salary or wages of a job. The salary given to someone must be in accordance with the criteria of the worker. Then we need a Machine Learning model to predict a person's salary. In this study, a classification model was made to determine a person to be categorized into salaries above 7 million and salaries below 7 million based on suitable criteria or attributes. This study uses the Python programming language and took 1000 samples from the dataset obtained from Kaggle. The Machine Learning method used is the Support Vector Machine. Then compared to the K-Nearest Neighbors method. In the SVM model the model accuracy was obtained of 87% and 86% for the KNN model. From the results of accuracy, it was found that the SVM model was better than the KNN model in conducting salary classifications based on existing jobs.