ACADEMIC YEAR: 2024-25
DATE:
MINI PROJECT REGISTRATION FORM
PROJECT BATCH NUMBER CSD_BATCH_12
S.N Hall Ticket Mobile
Name of the Student E-Mail ID
Number Number
TOPIC PROPOSED FOR PROJECT WORK : DETECTION OF FAKE ONLINE JOB
RECRUITMENT USING MACHINE LEARNING ALGORITHMS
NAME OF THE GUIDE WITH DESIGNATION:[Link](ASSOCIATE
PROFESSOR)
Signature of the Student
DEPARTMENT USE
1. Approved
2. Not Approved
Supervisor Head of the Department
BATCH NUMBER:12
DATE:
TITLE: Detection Of Fake Online Job Recruitment Using Machine Learning
Algorithms.
ABSTRACT
This paper presents a review on the proliferation of online job recruitment platforms has led to an
increase in fraudulent activities, with fake job postings deceiving unsuspecting job seekers. Detecting these
fraudulent job postings manually is time-consuming and often ineffective. Hence, leveraging machine learning
algorithms becomes imperative to automate the process and enhance detection accuracy. In this study, we
propose a framework for detecting fake online job recruitment using various machine learning techniques. Fake
online job recruitment scams have become increasingly prevalent, posing significant risks to individuals and
organizations. These scams exploit the anonymity of the internet to deceive job seekers with fraudulent offers,
often leading to financial or personal data loss. This study leverages machine learning techniques to detect and
mitigate fake online job recruitment activities. By analyzing a dataset of job postings, we identify key linguistic,
behavioral, and structural patterns indicative of fraudulent [Link] collect a dataset comprising genuine and
fake job postings, incorporating textual, visual, and metadata features. Subsequently, we preprocess the data and
engineer features to represent the characteristics of both genuine and fake postings effectively. We then train
and evaluate several machine learning models, including support vector machines, random forests, and deep
learning architectures, to classify job postings as genuine or fake. Furthermore, we conduct extensive
experiments to compare the performance of these models in terms of accuracy, precision, recall, and F1-score.
This approach not only enhances the security of online job platforms but also provides a scalable solution to
protect users from cyber threats. Future research could focus on real-time fraud detection and integrating this
system into existing job portals for proactive threat management. Our results demonstrate the efficacy of
machine learning algorithms in accurately identifying fake online job recruitment, thereby providing a valuable
tool for job seekers and recruitment platforms to mitigate the risks associated with fraudulent activities.
EXISTING SYSTEM:
In the existing system, the detection of fake online job recruitment typically relies on manual intervention,
where human moderators or administrators manually review job postings to identify suspicious or fraudulent
activities. This process is time-consuming, labor-intensive, and prone to errors, as human reviewers may
overlook subtle indicators of fraud or may not be able to handle the large volume of job postings effectively.
Additionally, the existing systems may lack comprehensive features or algorithms capable of detecting
sophisticated forms of fraudulent job postings, leading to increased risks for job seekers and potential financial
losses for recruitment platforms.
DRAWBACKS OF EXISTING SYSTEM:
● Limited Detection Capabilities: Traditional systems often rely on manual moderation or basic rule-
based algorithms, making them ineffective against sophisticated scams that evolve over time.
● High False Positives: Existing systems may inaccurately flag legitimate job postings as fraudulent,
frustrating employers and undermining trust in the platform.
● Lack of Real-Time Analysis: Many current systems cannot provide instantaneous detection, allowing
fake job postings to remain active long enough to target unsuspecting users.
PROPOSED SYSTEM:
● In the proposed system, we aim to develop an automated solution for detecting fake online job
recruitment using machine learning algorithms. The proposed system will leverage various machine
learning techniques to analyze textual, visual, and metadata features extracted from job postings to
identify patterns indicative of fraudulent activities. By incorporating advanced natural language
processing (NLP) algorithms, computer vision techniques, and feature engineering methodologies, the
proposed system will be capable of accurately distinguishing between genuine and fake job postings in
real-time. Moreover, the system will provide flexibility for integration with existing job recruitment
platforms, enabling seamless deployment and scalability. Overall, the proposed system offers significant
advantages over the existing manual approaches by providing a more efficient, accurate, and scalable
solution for detecting fake online job recruitment, thereby enhancing the trust and reliability of online
job marketplaces for both job seekers and recruiters.
ADVANTAGES OF PROPOSED SYSTEMS:
1. Efficiency and Speed
● Automated Detection: The system can analyze thousands of job postings in real-time, significantly
reducing the time required compared to manual review.
● Scalability: Capable of handling large volumes of data from multiple platforms simultaneously, making
it suitable for high-traffic job portals.
2. Improved Accuracy
● Advanced Algorithms: By leveraging machine learning and NLP, the system can detect subtle patterns
and inconsistencies often missed by human reviewers.
● Reduced Human Error: Automation eliminates biases and fatigue that can affect manual evaluation.
3. Cost-Effectiveness
● Lower Operational Costs: Reduces the need for extensive manpower to manually screen job postings,
saving costs in the long run.
● Prevention of Losses: Minimizes the financial and reputational damage caused by fraudulent job
postings.
4. Real-Time Fraud Detection
● Proactive Measures: Detects and flags fake job postings instantly, allowing platforms to take timely
actions before users are affected.
5. Enhanced Trust and Reliability
● User Confidence: Builds trust among job seekers and recruiters by ensuring a safer online environment.
● Platform Reputation: Improves the credibility of job recruitment platforms by reducing the prevalence
of scams.
SOFTWARE REQUIREMENTS:
Operating system : Windows 11
Coding Language : Python
HARDWARE REQUIREMENTS:
❖ System : Intel Core 2 Dual
❖ Hard Disk : 160 GB
❖ Monitor : 15’’ LED
❖ Input Devices : Keyboard, Mouse
❖ Ram : 32 GB
