The research question addressed in this study is to analyze the evolution of software quality models, and to investigate the similarities between the most known quality models. The models are decomposed according to their inherent structure, then are analyzed based on their subcharacteristics, at several levels. The focus of the study is on the lexical analysis of software quality models based on natural language processing, with the purpose to highlight the similarities in software quality models. In the end, the current software quality model ISO25010 is compared with all previous ones.