Classification of Fake Profiles Features in Facebook using K-Nearest Neighbour (KNN), Neural Network (NN) and Support Vector Machine (SVM)
Today, people rely heavily on Online Social Networks (OSN) which have piqued the interest of cyber criminals to carry out malicious acts. Coupled with the existence of illegal companies that make transactions with fake account services. To cater this fake account problem in OSN, this study focuses o...
Saved in:
Main Author: | |
---|---|
Format: | Thesis |
Language: | English English |
Published: |
2021
|
Subjects: | |
Online Access: | http://ir.upnm.edu.my/id/eprint/250/1/CLASSIFICATION%20OF%20FAKE%20%2825p%29.pdf http://ir.upnm.edu.my/id/eprint/250/2/CLASSIFICATION%20OF%20FAKE%20%28Full%29.pdf |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
id |
my-upnm-ir.250 |
---|---|
record_format |
uketd_dc |
spelling |
my-upnm-ir.2502023-06-15T01:26:22Z Classification of Fake Profiles Features in Facebook using K-Nearest Neighbour (KNN), Neural Network (NN) and Support Vector Machine (SVM) 2021-10-14 Ahmad Nasir, Ahmad Nazren Hakimi QA75 Electronic computers. Computer science Today, people rely heavily on Online Social Networks (OSN) which have piqued the interest of cyber criminals to carry out malicious acts. Coupled with the existence of illegal companies that make transactions with fake account services. To cater this fake account problem in OSN, this study focuses on identifying the most widely used fake Facebook accounts in OSN. It is included for their behaviour and features of fake account. This research methodology begins with data collection, identification of training and classifier functions, and finally validation and verification. The first process is to collect information about real and fake Facebook accounts. The second process is to use Facebook user feed data to understand user profile activity and identify the full set of features that play an important role in differentiating fake users from real Facebook users. Finally, we use these functions and identify the most important classifiers based on machine learning, which is mapping the identification of a total of 3 classifiers, namely K-nearest neighbour (KNN), support vector machine (SVM), and neural network (NN). The findings have revealed that the series of result shown that prediction the fake profile with good value of Classifying Accuracy (CA) which is KNN is 92%, NN is 94% and SVM is 95%. Same goes to Area Under Curve (AUC) which is KNN is 97%, NN is 98% and SVM is 98%. Ultimately, this finding will provide a new endeavour for countermeasure and protection of OSN users. 2021-10 Thesis http://ir.upnm.edu.my/id/eprint/250/ http://ir.upnm.edu.my/id/eprint/250/1/CLASSIFICATION%20OF%20FAKE%20%2825p%29.pdf text en public http://ir.upnm.edu.my/id/eprint/250/2/CLASSIFICATION%20OF%20FAKE%20%28Full%29.pdf text en validuser mphil masters Universiti Pertahanan Nasional Malaysia Centre For Graduate Studies |
institution |
Universiti Pertahanan Nasional Malaysia |
collection |
UPNM Institutional Repository |
language |
English English |
topic |
QA75 Electronic computers Computer science |
spellingShingle |
QA75 Electronic computers Computer science Ahmad Nasir, Ahmad Nazren Hakimi Classification of Fake Profiles Features in Facebook using K-Nearest Neighbour (KNN), Neural Network (NN) and Support Vector Machine (SVM) |
description |
Today, people rely heavily on Online Social Networks (OSN) which have piqued the interest of cyber criminals to carry out malicious acts. Coupled with the existence of illegal companies that make transactions with fake account services. To cater this fake account problem in OSN, this study focuses on identifying the most widely used fake Facebook accounts in OSN. It is included for their behaviour and features of fake account. This research methodology begins with data collection, identification of training and classifier functions, and finally validation and verification. The first process is to collect information about real and fake Facebook accounts. The second process is to use Facebook user feed data to understand user profile activity and identify the full set of features that play an important role in differentiating fake users from real Facebook users. Finally, we use these functions and identify the most important classifiers based on machine learning, which is mapping the identification of a total of 3 classifiers, namely K-nearest neighbour (KNN), support vector machine (SVM), and neural network (NN). The findings have revealed that the series of result shown that prediction the fake profile with good value of Classifying Accuracy (CA) which is KNN is 92%, NN is 94% and SVM is 95%. Same goes to Area Under Curve (AUC) which is KNN is 97%, NN is 98% and SVM is 98%. Ultimately, this finding will provide a new endeavour for countermeasure and protection of OSN users. |
format |
Thesis |
qualification_name |
Master of Philosophy (MPhil) |
qualification_level |
Master's degree |
author |
Ahmad Nasir, Ahmad Nazren Hakimi |
author_facet |
Ahmad Nasir, Ahmad Nazren Hakimi |
author_sort |
Ahmad Nasir, Ahmad Nazren Hakimi |
title |
Classification of Fake Profiles Features in Facebook using K-Nearest Neighbour (KNN), Neural Network (NN) and Support Vector Machine (SVM) |
title_short |
Classification of Fake Profiles Features in Facebook using K-Nearest Neighbour (KNN), Neural Network (NN) and Support Vector Machine (SVM) |
title_full |
Classification of Fake Profiles Features in Facebook using K-Nearest Neighbour (KNN), Neural Network (NN) and Support Vector Machine (SVM) |
title_fullStr |
Classification of Fake Profiles Features in Facebook using K-Nearest Neighbour (KNN), Neural Network (NN) and Support Vector Machine (SVM) |
title_full_unstemmed |
Classification of Fake Profiles Features in Facebook using K-Nearest Neighbour (KNN), Neural Network (NN) and Support Vector Machine (SVM) |
title_sort |
classification of fake profiles features in facebook using k-nearest neighbour (knn), neural network (nn) and support vector machine (svm) |
granting_institution |
Universiti Pertahanan Nasional Malaysia |
granting_department |
Centre For Graduate Studies |
publishDate |
2021 |
url |
http://ir.upnm.edu.my/id/eprint/250/1/CLASSIFICATION%20OF%20FAKE%20%2825p%29.pdf http://ir.upnm.edu.my/id/eprint/250/2/CLASSIFICATION%20OF%20FAKE%20%28Full%29.pdf |
_version_ |
1813223593386967040 |