Fake Job Prediction Using Machine Learning
V. Anbarasu, S. Selvakani, K. Vasumathi · International Journal of Darshan Institute on Engineering Research and Emerging Technologies · 2024
In order to mitigate the proliferation of deceptive employment solicitations on the internet, a sophisticated automated tool employing machine learning-based classification methodologies is posited within the confines of this scholarly work.Various classifiers are deployed to scrutinize online postings for fraudulent employment opportunities, and the outcomes of these classifiers are systematically juxtaposed to ascertain the most efficacious model for detecting spurious job listings.This approach facilitates the identification and subsequent elimination of counterfeit job posts from an extensive array of online submissions.The investigation encompasses two principal categories of classifiers: individual classifiers and ensemble classifiers, both instrumental in discerning deceitful job postings.Nonetheless, empirical findings unequivocally affirm that ensemble classifiers exhibit superior efficacy in discerning scams when compared to their singular counterparts.The technological landscape has ascended to a heightened echelon, ushering in a paradigm wherein corporations engage in the recruitment of personnel through the conduit of online methodologies.This not only expedites the acquisition of requisite personnel for businesses but also augurs well in terms of cost-effectiveness.The virtual expanse facilitates individuals in procuring employment commensurate with their qualifications and desired occupational spheres.However, the veracity of these posted job opportunities remains shrouded, posing an inherent challenge for job seekers.In response to this predicament, we proffer a pioneering software meticulously crafted to prognosticate the authenticity of job posts, discerning between genuine and spurious listings.Embarking upon the realm of machine learning, our innovative system, aptly named "Fake Job Post Prediction," leverages the formidable Random Forest classifier.This sophisticated algorithm boasts a commendable efficiency in generating precise outcomes, with a remarkable 98% accuracy vis-à-vis its predecessors.Recognizing the perils faced by students or job seekers navigating the labyrinth of online employment opportunities, our system becomes a beacon of protection against unwittingly submitting personal information to fraudulent job posts.Instances of potential deception, such as solicitation of application fees or promises of employment contingent upon monetary transactions, are thus preemptively averted through the discerning capabilities of our framework, thereby safeguarding users from falling prey to scams.