Suspicious online car advertisement detector using principle component analysis (PCA) / Fathiah Husna Firdaus

Firdaus, Fathiah Husna (2019) Suspicious online car advertisement detector using principle component analysis (PCA) / Fathiah Husna Firdaus. Degree thesis, Universiti Teknologi MARA (UiTM).
Abstract

E-shopping or also known as Electronic Commerce (E-Commerce) is an online shopping website that have been evolved in these past years. The evolution of online shopping website has been beneficial to both of the sellers and customers mainly in terms of time and cost. Despite of the goods of online shopping, there is possibility that the advertisement in the website is considered suspicious and a potential scam. Hence, this study is to propose suspicious advertisement detector by using data mining technique which is Principal Component Analysis (PCA). PCA is a dimensionality reduction technique that is robust and ease the visualization task. The result of PCA will be used to identify the outliers from the PCA scatter plot. The further analysis is done by apply statistical box plot method, K-Means Clustering and compare the distance of outliers and its centroid. The outliers from the result are the suspicious advertisements which could be potential scammer or genuine seller that differs in value of price and mileage of a car in the advertisement.

Item Details
Edit Item
Edit Item
Downloads & Files
[thumbnail of 87224.pdf]
Text
87224.pdf
Download (7MB)
Location & Physical Holdings
Digital Copy
Physical Location
  • Koleksi Akses Terhad | PTAR Utama | Shah Alam
Item Status
On Shelf
Indexing & Metrics