Comparative Study of Web Page Classification Approaches

Vinod, Pooja; Prajapati, Purvi

doi:10.5120/ijca2018916994

Cited by 5 publications

(1 citation statement)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For instance, subject classification would be determining whether a page is about arts, business, or sports. [5]. • Functional classification takes into consideration the function that the website page serves.…”

Section: Webpages Classification Typesmentioning

confidence: 99%

Webpage Classification Using Ensemble Machine Learning

Altaee

2023

IJICI

View full text Add to dashboard Cite

These days, it is not easy to get the correct information after typing a keyword into a search engine because so many results are returned. Classification of Web pages is a technique that helps us locate the wanted information quickly and effectively. In addition, website categorization is crucial for businesses that provide marketing and analytical solutions because it enables them to create a well-balanced mix of search engine and directory listings. This will give marketers a better idea of where their local company listings appear online, allowing them to have more judgment about initiative and strategy. Therefore, the research aimed to construct a classification system based on a dataset of English web pages. This information has been acquired from the Kaggle website and consisted of 1408 distinct rows organized into 16 categories. The research has employed mixed strategies to determine which strategy for Web page categorization would yield the best results. The first strategy puts into practice a collection of machine-learning algorithms. It assesses how well they accomplish the given classification task. Ensemble stacking is the second strategy, and it is employed to enhance the classification of websites. Comparing the results of the two strategies reveals that Ensemble stacking, the second strategy, was the more influential architecture for classifying web pages this approach had 0.95 F1-score, 0.95 accuracy, 0.95 precision, and 0.95 recall achieved by this method. The first approach, which made use of machine learning techniques, on the other hand, received an F1-score of 0.93, 0.94 for precision, 0.93 for recall, and 0.93 for accuracy.

show abstract

Section: Webpages Classification Typesmentioning

confidence: 99%