Scribd
Upload
593 views

A Comparative Study On Fake Job Post Prediction Using Different Data Mining Techniques

This paper proposes using different data mining techniques and machine learning classifiers like KNN, decision trees, support vector machines, naive Bayes, random forest, multilayer perceptron, and deep neural networks to predict if a job post is real or fraudulent. The researchers experimented on the Employment Scam Aegean Dataset containing 18,000 samples and found that a deep neural network achieved 98% accuracy. Existing methods are discussed that used classifiers like random forest and SVM on this dataset with accuracies up to 97.4%. The proposed system uses this same dataset but converts attributes to categorical values before classifying with machine learning to detect fake job posts.

Uploaded by

bits computers
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
593 views

A Comparative Study On Fake Job Post Prediction Using Different Data Mining Techniques

This paper proposes using different data mining techniques and machine learning classifiers like KNN, decision trees, support vector machines, naive Bayes, random forest, multilayer perceptron, and deep neural networks to predict if a job post is real or fraudulent. The researchers experimented on the Employment Scam Aegean Dataset containing 18,000 samples and found that a deep neural network achieved 98% accuracy. Existing methods are discussed that used classifiers like random forest and SVM on this dataset with accuracies up to 97.4%. The proposed system uses this same dataset but converts attributes to categorical values before classifying with machine learning to detect fake job posts.

Uploaded by

bits computers
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 5

A Comparative Study on Fake Job Post Prediction

Using Different Data mining Techniques

ABSTRACT

In recent years, due to advancement in modern technology and social


communication, advertising new job posts has become very common issue in the
present world. So, fake job posting prediction task is going to be a great concern for
all. Like many other classification tasks, fake job posing prediction leaves a lot of
challenges to face. This paper proposed to use different data mining techniques and
classification algorithm like KNN, decision tree, support vector machine, naive
bayes classifier, random forest classifier, multilayer perceptron and deep neural
network to predict a job post if it is real or fraudulent. We have experimented on
Employment Scam Aegean Dataset (EMSCAD) containing 18000 samples. Deep
neural network as a classifier, performs great for this classification task. We have
used three dense layers for this deep neural network classifier. The trained classifier
shows approximately 98% classification accuracy (DNN) to predict a fraudulent job
post.

EXISTING SYSTEM

Many researches occurred to predict if a job post is real or fake. A good number of
research works are to check online fraud job advertiser. Vidros [1] et al. identified
job scammers as fake online job advertiser. They found statistics about many real
and renowned companies and enterprises who produced fake job advertisements or
vacancy posts with ill-motive. They experimented on EMSCAD dataset using
several classification algorithms like naive bayes classifier, random forest classifier,
Zero R, One R etc. Random Forest Classifier showed the best performance on the
dataset with 89.5% classification accuracy. They found logistic regression
performing very poor on the dataset. One R classifier performed well when they
balanced the dataset and experimented on that. They tried in their work to find out
the problems in ORF model (Online Recruitment Fraud) and to solve those problems
using various dominant classifiers.

Alghamdi [2] et al. proposed a model to detect fraud exposure in an online


recruitment system. They experimented on EMSCAD dataset using machine learning
algorithm. They worked on this dataset in three steps- data pre-processing, feature
selection and fraud detection using classifier. In the preprocessing step, they
removed noise and html tags from the data so that the general text pattern remained
preserved. They applied feature selection technique to reduce the number of
attributes effectively and efficiently. Support Vector Machine was used for feature
selection and ensemble classifier using random forest was used to detect fake job
posts from the test data. Random forest classifier seemed a tree structured classifier
which worked as ensemble classifier with the help of majority voting technique. This
classifier showed 97.4% classification accuracy to detect fake job posts.

Huynh [3] et al. proposed to use different deep neural network models like Text
CNN, Bi-GRU-LSTM CNN and Bi- GRU CNN which are pre-trained with text
dataset. They worked on classifying IT job dataset. They trained IT job dataset on
TextCNN model consisting of convolution layer, pooling layer and fully connected
layer. This model trained data through convolution and pooling layers. Then the
trained weights were flattened and passed to the fully connected layer. This model
used softmax function for classification technique. They also used ensemble
classifier (Bi-GRU CNN, Bi-GRULSTM CNN) using majority voting technique to
increase classification accuracy. They found 66% classification accuracy using
TextCNN and 70% accuracy for Bi-GRU- LSTM CNN individually. This
classification task performed best with ensemble classifier having an accuracy of
72.4%.

Zhang [4] et al. proposed an automatic fake detector model to distinguish between
true and fake news (including articles, creators, subjects) using text processing. They
had used a custom dataset of news or articles posted by PolitiFact website twitter
account. This dataset was used to train the proposed GDU diffusive unit model.
Receiving input from multiple sources simultaneously, this trained model performed
well as an automatic fake detector model.

Disadvantages

1) The system is implemented by Conventional Machine Learning.


2) The system doesn’t implement for analyzing large data sets.

PROPOSED SYSTEM

The system has used EMSCAD to detect fake job post. This dataset contains 18000
samples and each row of the data has 18 attributes including the class label. The
attributes are job_id, title, location, department, salary_range, company_profile,
description, requirements, benefits, telecommunication, has_company_logo,
has_questions, employment_type, required_experience, required_education,
industry, function, fraudulent (class label). Among these 18 attribute, we have used
only 7 attributes which are converted into categorical attribute. T elecommuting,
has_company_logo, has_questions, employment_type, required experience,
required_education and fraudulent are changed into categorical value from text
value. For example, “employment_type” values are replaced like this- 0 for “none”,
1 for ‘full-time”, 2 for “part-time” and 3 for “others”, 4 for “contract’ and 5 for
“temporary”. The main goal to convert these attributes into categorical form is to
classify fraudulent job advertisements without doing any text processing and natural
language processing. In this work, we have used only those categorical attributes.
Advantages
1) The proposed has been implemented EMSCAD technique which is very accurate
and fast.
2) The system is very effective due to accurate detection of Fake job posts which
creates inconsistency for the job seeker to find their preferable jobs causing a huge
waste of their time.

SYSTEM REQUIREMENTS

➢ H/W System Configuration:-

➢ Processor - Pentium –IV


➢ RAM - 4 GB (min)
➢ Hard Disk - 20 GB
➢ Key Board - Standard Windows Keyboard
➢ Mouse - Two or Three Button Mouse
➢ Monitor - SVGA

SOFTWARE REQUIREMENTS:

 Operating system : Windows 7 Ultimate.

 Coding Language : Python.

 Front-End : Python.

 Back-End : Django-ORM

 Designing : Html, css, javascript.

 Data Base : MySQL (WAMP Server).

You might also like