THE DETECTION OF FAKE JOB POSTS BY USING K-NEAREST NEIGHBOR (KNN)

Abstract

Every day, there are issues on many job-calling websites on the internet. Some of the jobs advertised online are actually fake jobs that lead to the theft of sensitive information. So it needs to be identified. Using the term frequency inverse document frequency (TF-IDF) method and the K-Nearest Neighbor (KNN) algorithm, the system detects fake job postings. TF-IDF is one of the statistical techniques for calculating the importance or score of each word in a document. In order to extract features, TF-IDF is employed. The purpose of this system is to classify real or fake jobs by using the KNN classifier. This system is implemented using 17866 jobs as the Fake Job Posts detection dataset. The accuracy of the proposed system is measured by using a confusion matrix (precision, recall, and F-Measure). The experiment results have been many times used with the existing actual data by using K-Nearest Neighbor algorithm with K value changes (K=1, 3, 5, 7, 9). According to the comparison results, the proposed system has achieved high accuracy many times.