Software requirements are mostly expressed in natural language, rendering them unstructured and thus not readily machine-understandable. As a result, a number of challenges have arisen in Requirements Engineering (RE), especially with respect to tasks such as requirements analysis, modelling, and management, which can be accomplished much more efficiently with automated support. These,however, can be alleviated by adding structure to software requirements by attaching machine-readable semantic metadata, i.e., annotations, that captures meaning of a written requirements statement. FrameNet provides a rich set of semantic frames that can be leveraged for this purpose. In this study, we investigate how FrameNet frames can be adopted to annotate software requirements. As part of this work, we proposed a semi-automatic annotation approach and produced a novel corpus of requirements documents annotated with FrameNet semantic frames. Our evaluation of annotations provided by our two annotators showed 'substantial' agreement between them, implying that the resulting corpus can be considered as reliable.
@InProceedings{ALHOSHAN18.5, author = {Waad Alhoshan ,Riza Batista-navarro and Liping Zhao}, title = {A FrameNet-based Approach for Annotating Software Requirements.}, booktitle = {Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018)}, year = {2018}, month = {may}, date = {7-12}, location = {Miyazaki, Japan}, editor = {Tiago Timponi Torrent and Lars Borin and Collin F. Baker}, publisher = {European Language Resources Association (ELRA)}, address = {Paris, France}, isbn = {979-10-95546-04-7}, language = {english} }
Summary of the paper