Siti Hawa, Apandi and Jamaludin, Sallim and Rozlina, Mohamed and Norkhairi, Ahmad (2024) Data Pre-processing of Website Browsing Records: To Prepare Quality Dataset for Web Page Classification. International Journal on Informatics Visualization, 8 (1). pp. 239-246. ISSN 2549-9904. (Published)
|
Pdf
Data Pre-processing of Website Browsing Records - To Prepare Quality Dataset for Web Page Classification.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (3MB) | Preview |
Abstract
The increased usage of the internet worldwide has led to an abundance of web pages designed to supply information to internet users. The use of web page classification is becoming increasingly necessary to organize the growing number of web pages. This classification model serves as a tool to restrict internet usage to specific categories of web pages. To develop the classification model, it’s crucial to check the quality of the dataset, as it determines the performance of the web page classification model. Raw datasets are typically unreliable and subject to noise, which complicates data analysis. This is why data pre-processing is necessary to prepare the dataset properly. In this study, website browsing records serve as the dataset. The primary goal of this paper is to investigate data pre-processing techniques for website browsing records, focusing on Game and Online Video Streaming web pages. Data pre-processing involves two main steps: data cleaning and web content pre-processing. After completing the data cleaning process, the datasets are reduced from the original. This demonstrates that many datasets can be eliminated due to their inactivity or unsuitability as the datasets for Game and Online Video Streaming web pages. Meanwhile, web content pre-processing removes noise from an HTML document, retaining only relevant words that can represent the web page by creating a word cloud image. Convolutional Neural Networks (CNN) will be used to construct a model for categorizing web pages to determine whether they fall under Game or Online Video Streaming. The pre-processed data will be used as the input for this model.
| Item Type: | Article |
|---|---|
| Additional Information: | Indexed by Scopus |
| Uncontrolled Keywords: | Data cleaning; data pre-processing; web content pre-processing; web page classification; website browsing records |
| Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software |
| Faculty/Division: | Institute of Postgraduate Studies Faculty of Computing |
| Depositing User: | Miss Amelia Binti Hasan |
| Date Deposited: | 09 May 2024 00:00 |
| Last Modified: | 09 May 2024 00:00 |
| URI: | http://umpir.ump.edu.my/id/eprint/40046 |
| Download Statistic: | View Download Statistics |
Actions (login required)
 |
View Item |