
370 Empirical Research in Software Engineering
system and, therefore, these qualities should be incorporated in the architectural design
as early as possible, which is not the case. Hence, in this example, we intend to employ
text mining techniques to mine these requirements and convert them into a structured
form that can then be used for the prediction of the unknown requirements into their
respective nonfunctional qualities. The data presents few requirements extracted from
promise data repository. Here, we categorize the requirements into three different type
of NFRs, namely, availability (A), look-and-feel (LF), and usability (U). These ...