Computational analysis of web page metrices for search engine optimization

Abstract

The relevance of a web page (based on search keywords) is a naturally biased topic that is entirely dependent on the knowledge, interests, and attitude of the reader. Any web page is reliant on these elements. The fast growth of the internet is one element that makes it challenging for search engines to provide relevant results to consumers in a timely manner. Search engines employ classified directories to store webpages, and some search engines even rely on human expertise in this process. The majority of web pages classify their content using automated algorithms. This paper proposes a novel technique to analyzing SEO attributes of top 5 sites in google (based on keywords search on google) and making predictions about how to use suitable SEO attributes to improve site ranking on google search using machine learning algorithms based on experts' experience. The findings of the experiments suggest that machine learning may be used to anticipate the degree of web page adaptation to SEO advice. The proposed approach has practical value in that it provides the foundation for developing software agents and expert systems that can automatically detect web pages, or parts of web pages, that need to be improved in order to comply with SEO guidelines and, as a result, gain higher search engine rankings. The findings of this study also contribute to the topic of determining the best values for ranking criteria used by search engines to rank web pages. Experiments in this study reveal that the page title, meta description, H1 tag (header), and body text are essential aspects to consider when creating a web page, which is consistent with earlier studies. Another outcome of this study is a new data set based on machine learning web page prediction that can be used in future studies.