EA-Miner: Towards Automation in Aspect-Oriented Requirements Engineering - Transactions on Aspect-Oriented Software Development - page 17

Figure 7 shows the viewpoints details page with the list of viewpoints on the top
left side. When the user selects a viewpoint, the requirements of this viewpoint are
displayed on the right and its crosscutting relationships are displayed in the view on
the bottom.
Enables the user to add, delete and
update viewpoints and their
requirements
Fig. 7. Viewpoints identified and their requirements and crosscutting relationships
The user also has the option of adding/deleting/updating requirements or
viewpoints. As mentioned before, EA-Miner lists candidate concepts and the
requirements engineer is able to update the model built by the tool. Guidelines are
available for informing the user to make a good decision. Section 4.3 discusses these
guidelines.
4.1.3 Early Aspect Identification, Presentation and Structuring (Activities 2.3
and 3.2)
The rules used for identifying non-functional early aspects are based on the semantic
tag attributed to a word as discussed in Sect. 3.1.1. The identification approach for the
early aspects (see Figs. 8, 10) works similar to what was described for viewpoints.
The tool looks for the semantic tag of each word in the sentences of the parsed input
document and searches a lexicon 2 file to check if there is one entry of the same word
with the same semantic tag. The word does not need to be exactly equal as a stemmer
is used to reduce words to the base form when comparing the words (e.g.,
“authorised”, “authorises” and “authorise” are reduced to the same base form).
2 An excerpt of the lexicon file in shown in Fig. 8. It contains entries of NFR catalogue type
words [14] along with their semantic tags.