An improved approach for web page quality assessment

Internet websites need to be measured and evaluated for quality and better understanding. Several Metrics were proposed to correspond with items that Web usability guideline associate with good design such as word count, total page size in bytes, body text percentage, average link text count and others. This study investigates empirically the web page quality on the basis of the 16 assessed metrics. Support Vector Machine (SVM) model is proposed to predict the good web pages and not good web pages. Web sites are collected from Webby Awards data (2001-2010) and Top Ten PC Magazines. The findings of quantitative analysis of web page attributes are expressed and how these attributes are calculated. The metrics captured in SVM model can be used to predict the website designs the good and the bad.