MODEL-DRIVEN WEB USAGE ANALYSIS FOR THE EVALUATION OF WEB APPLICATION QUALITY

Abstract

So far, conceptual modeling of Web applications has been used primarily in the upper part of the life cycle, as a driver for system analysis. Little attention has been put on exploiting the conceptual speci¯cations developed during analysis for application evalu- ation, maintenance and evolution. This paper illustrates an approach for integrating the use of conceptual models in the lower part of the application life cycle. The approach is based on the adoption of conceptual logs, which are Web usage logs enriched with meta-data deriving from the application conceptual speci¯cations. In particular, the pa- per illustrates how conceptual logs are generated and exploited in Web usage evaluation and mining, so as to achieve a deeper and systematic quality evaluation of Web applica- tions. A prototype tool supporting the generation of conceptual logs and the evaluation activities is also presented.