TAPHSIR

TAPHSIR is a tool associated with the ICSE22 technical paper -- titled "Automated Handling of Anaphoric Ambiguity in Requirements: A Multi-solution Study", for handling anaphoric ambiguity in natural-language requirements. TAPHSIR provides a hybrid solution composed of two components. The first component employs machine learning (ML) for detecting anaphoric ambiguity. The second component builds on the recent advances in natural language processing, specifically the re-trained language model SpanBERT, for providing anaphora resolutions (when possible).

The tool was developed at SnT / University of Luxembourg with funding from Luxembourg's National Research Fund (FNR).