Anomaly detection in Web applications: a review of already conducted case studies

The quality of Web applications is everyday more important. Web applications are crucial vehicles for commerce, information exchange, and a host of social and educational activities. Since a bug in a Web application could interrupt an entire business and cost millions of dollars, there is a strong demand for methodologies, tools and models that can improve the Web quality and reliability. Aim of our ongoing-work has been to investigate, define and apply a variety of analysis and testing techniques able to support the quality of Web applications. Validity of our solutions was assessed by extensive empirical work. A critical review of this five year long work shows that only 40% of the randomly selected real-world Web applications exhibit no anomalies/failures. Some tables reported in this paper summarize the relations between type of anomalies found and analyses applied. We are in need of better methodologies, techniques and tools for developing, maintaining and testing Web applications.