Efficient Streamlined Online Arabic Web Page Classification Using Artificial Bee Colony Optimization

Abstract

In the realm of online information retrieval, Arabic web content presents unique challenges due to the complexity of the Arabic language and the varying quality of available materials. These complexities often confound search engines, hindering precise web page classification. This research addresses these challenges with a streamlined approach to effective online Arabic web page classification. Our strategy involves two key components: First, we employ a comprehensive text mining approach to extract valuable insights from a corpus of documents. This begins with text extraction from the online web page after providing the URL and the removal of stop words to enhance data quality. Additionally, we leverage Natural Language Processing (NLP), specifically lemmatization, to normalize text and reduce linguistic variations, ensuring consistent and meaningful representation. To complete the text mining process, we use the Bag of Words (BoW) model to transform preprocessed text data into a numerical format, capturing word frequencies. Second, we harness the power of the Artificial Bee Colony (ABC) optimization algorithm, inspired by bees' foraging behavior, as a pivotal element in our decision-making process. This algorithm provides a robust framework for optimizing classification tasks. Empirical results affirm the effectiveness of our approach, achieving an impressive 95.349% accuracy rate. This advancement bridges the gap between the intricacies of the Arabic language and efficient web content organization, promising a more informative Arabic web landscape.