The Webis-Mnemonics-17 corpus is a collection of 1048 human-chosen sentences for password generation and memorization (so-called mnemonics). It is designed to test hypotheses on statistical properties of such mnemonics.
To download the corpus use the following link:
(2.2 MB, sha256 sum: 07509754f2a831acc420a8a2686336ccb8708a6e1df24c5633d0c49a3a5b2331)
If you use the dataset in your research, please send us a copy of your publication. We kindly ask you to refer to the corpus via [bib].
The corpus was collected using a 4-step study: sentence creation; sentence and password memorization; background questions; recall.
In addition to the raw mnemonics, the corpus comes with detailed logs of the study participants' interactions.
The corpus is described in more detail in this publication: [bib]
Students: Norina Marie Grosch