LSTM Based Approach For Detecting Fake Job Posting - An Empirical study

Sinchana N S; N. Kumaresh

doi:10.36948/ijfmr.2025.v07i05.58663

LSTM Based Approach For Detecting Fake Job Posting - An Empirical study

Author(s)	Ms. Sinchana N S, Dr. N. Kumaresh
Country	India
Abstract	In today’s digital job market, the proliferation of fraudulent job postings poses a significant threat to job seekers, leading to financial loss, identity theft, and emotional distress. Traditional detection methods often fail to scale or accurately identify the subtle patterns in modern scams. To combat this, we present an intelligent, hybrid system designed to unmask fraudulent job listings. Our methodology began with a comparative study to identify the most effective classification technology, evaluating tra-ditional machine learning models (Random Forest, XGBoost, Light GBM) against a Long Short-Term Memory (LSTM) deep learning network. A critical challenge was the severe class imbalance in our dataset. To ensure a fair comparison, we applied the Synthetic Minority Over-sampling Technique (SMOTE) to the training data for the classical models and used class weights for the LSTM. Our ex-periments concluded that the LSTM model delivered the most reliable classifications for the minority (fraudulent) class, due to its superior ability to capture the contextual nuances of language. Based on this data-driven selection, we implemented a final hybrid detection system centered on the LSTM. This system is enhanced by a real-time, rule-based keyword scanner that acts as a crucial first line of defense, immediately flagging posts containing high-risk scam phrases. The entire system is delivered as a user-friendly web application with separate, role-based dashboards for job seekers and administrators. This comprehensive approach confirms that a hybrid strategy, combining a carefully selected deep learning model with explicit, rule-based checks, can substantially improve the effective-ness, reliability, and usability of fake job detection tools, offering a practical solution to a pressing re-al-world problem.
Keywords	Fake job detection, Natural language processing, LSTM, Scam keyword flagging, Class Imbalance, Synthetic Minority Over-sampling Technique.
Field	Computer
Published In	Volume 7, Issue 5, September-October 2025
Published On	2025-10-23
DOI	https://doi.org/10.36948/ijfmr.2025.v07i05.58663

View / Download PDF File

E-ISSN 2582-2160

doi

CrossRef DOI is assigned to each research paper published in our journal.

IJFMR DOI prefix is
10.36948/ijfmr

Downloads

Research Paper Format Copyright Permission Form and Undertaking Form Cover Page Vol 8 Isu 2 Cover Page Vol 8 Isu 1 Cover Page Vol 7 Isu 6

All research papers published on this website are licensed under Creative Commons Attribution-ShareAlike 4.0 International License, and all rights belong to their respective authors/researchers.

CC-BY-SA

About IJFMR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9687-828-838	editor@ijfmr.com

International Journal For Multidisciplinary Research

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Bi-monthly Scholarly International Journal

LSTM Based Approach For Detecting Fake Job Posting - An Empirical study

Share this