Prediction of user behavior based on clustering techniques

Also available in printed version: ZA4235 K46 2015 raf

Bibliographic Details
Main Author: El-Khodary, Rehab Hassan
Other Authors: Salwani Mohd. Daud, supervisor
Format: Master's thesis
Language:English
Published: Universiti Teknologi Malaysia 2025
Subjects:
Online Access:https://utmik.utm.my/handle/123456789/60247
Abstract Abstract here
_version_ 1854975067344076800
author El-Khodary, Rehab Hassan
author2 Salwani Mohd. Daud, supervisor
author_facet Salwani Mohd. Daud, supervisor
El-Khodary, Rehab Hassan
author_sort El-Khodary, Rehab Hassan
description Also available in printed version: ZA4235 K46 2015 raf
format Master's thesis
id utm-123456789-60247
institution Universiti Teknologi Malaysia
language English
publishDate 2025
publisher Universiti Teknologi Malaysia
record_format dspace
record_pdf Abstract
spelling utm-123456789-602472025-08-21T00:05:24Z Prediction of user behavior based on clustering techniques El-Khodary, Rehab Hassan Salwani Mohd. Daud, supervisor Web usage mining Data mining Also available in printed version: ZA4235 K46 2015 raf The use of web technology has increased by a great extent in the recent times. Millions of users spend time surfing the internet to obtain information or for recreational activities leaving many untapped data behind that can be used in several commercial domains like mobile marketing, ecommerce websites, social networking sites and entrepreneur franchises. This creates new possibilities to communicate with the web users based on their historical data and target their interests and satisfaction. The overall aim of this research is to enhance the accuracy of predicting the user’s future behavioral patterns. Web Usage Mining is an important type of Web Mining, which deals with extraction of interesting knowledge from the web log files. There have been rigorous research works that done in this field but basically this research emphasize on the prediction of user behavior to target the users with personalized features. This research also reports comparisons between the web usage mining methods with introducing different applications, which gives the development overview of the research. A system is implemented using the web log files taken from UTM web server, all the data mining processes is applied practically. The pattern discovery process is applied in two different ways, the first one by applying the unsupervised k-means clustering on all users as one data set and the test results is about 79%. While the second one is done by integrating the unsupervised K-means clustering with the supervised Probabilistic Neural Network, this process is applied on every user data set individually. Then, the system is tested and yields a good results that reaches 82%. Finally, the system achieved a good accuracy level comparing with other systems. sof UTM 101 p. Thesis (Sarjana Sains (Kejuruteraan Sistem Komputer))-Universiti Teknologi Malaysia, 2015 2025-03-17T06:35:12Z 2025-03-17T06:35:12Z 2015 Master's thesis https://utmik.utm.my/handle/123456789/60247 valet-20170515-132937 vital:100304 ENG Closed Access UTM Complete Unpublished application/pdf Universiti Teknologi Malaysia
spellingShingle Web usage mining
Data mining
El-Khodary, Rehab Hassan
Prediction of user behavior based on clustering techniques
thesis_level Master
title Prediction of user behavior based on clustering techniques
title_full Prediction of user behavior based on clustering techniques
title_fullStr Prediction of user behavior based on clustering techniques
title_full_unstemmed Prediction of user behavior based on clustering techniques
title_short Prediction of user behavior based on clustering techniques
title_sort prediction of user behavior based on clustering techniques
topic Web usage mining
Data mining
url https://utmik.utm.my/handle/123456789/60247
work_keys_str_mv AT elkhodaryrehabhassan predictionofuserbehaviorbasedonclusteringtechniques