Tagged near-synonyms in requirements specifications

Published: 22-12-2018| Version 1 | DOI: 10.17632/yjnp5chzbv.1
Contributors:
Fabiano Dalpiaz,
F. Basak Aydemir

Description

The spreadsheet complements the paper "Detecting Terminological Ambiguity in User Stories: Tool and Experimentation" published in the Information & Software Technology Journal. It consists of a set of term couples that belong to 28 data sets, and that were considered as near-synonyms by some manual taggers. The spreadsheet has a single sheet with the following columns: - Dataset: an identifier from 1 to 28 that refers to the data set - Type: noun or verb - Term A: the first term - Term B: the second term - T1: a boolean value that states if the first manual tagger considered the terms as near-synonyms - T2: a boolean value that states if the second manual tagger considered the terms as near-synonyms - Gold set 1: a boolean value that indicates if the couple is in the first version of the gold set of near-synonyms, created by the data set owner based on the inputs of T1 and T2 - T3 (manual): a boolean value that states if the time-constrained, manual tagger indicated the two terms in that row as near-synonyms - T4 (REVV-Light): a boolean value that states if the time-constrained, REVV-Light supported tagger indicated the two terms in that row as near-synonyms - Gold set 2: a boolean value that indicates if the couple is in the first version of the gold set of near-synonyms, created by the data set owner based on all the other inputs

Files