ONLINE JOB PORTAL

Abstract

In recent times, with the rapid advancements in
modern technology and the widespread use of social
communication platforms, advertising new job
vacancies has become a common practice worldwide.
Consequently, detecting fraudulent job postings has
emerged as a significant concern. Similar to many
other classification tasks, predicting fake job postings
presents numerous challenges. This study proposes
the utilization of various machine learning techniques
and classification algorithms, including KNN,
decision trees, support vector machines, naive Bayes
classifier, random forest classifier, multi-layer
perceptron, and deep neural networks, to discern
whether a job posting is genuine or deceptive.To
combat fraudulent job postings on the internet, this
paper introduces an automated tool leveraging
machine learning-based classification techniques.
Multiple classifiers are employed to identify
fraudulent postings online, and the results from these
classifiers are compared to determine the most
effective employment scam detection model. This
approach aids in identifying fake job listings from a
vast array of postings. The study evaluates two
primary types of classifiers: single classifiers and
ensemble classifiers. While single classifiers, such as
those trained specifically for detecting fraudulent job
postings in West Bengal, exhibit promising results,
ensemble classifiers prove to be superior in detecting
scams.Instances of fraudulent job postings present an
ideal opportunity for fraudsters, particularly
exploiting the desperation induced by significant
tragedies. Consequently, many individuals fall victim
to these scams, where fraudsters seek to obtain
personal information such as addresses, financial
account details, and social security numbers from
their targets. As university students, we have
encountered numerous instances of these scam
emails, wherein fraudsters entice users with
seemingly lucrative job opportunities, only to
demand payment or personal information in return
for promised employment. Addressing this perilous
issue requires the application of natural language
processing (NLP) and machine learning
methodologies.