Analogy-based explanation evaluation dataset for HCOMP 2022 paper "It Is Like Finding a Polar Bear in the Savannah! Concept-level AI Explanations with Analogical Inference from Commonsense Knowledge."
DOI:10.4121/de1df5c0-5430-40f9-ba9a-ba0d1f415f28.v1
The DOI displayed above is for this specific version of this dataset, which is currently the latest. Newer versions may be published in the future.
For a link that will always point to the latest version, please use
DOI: 10.4121/de1df5c0-5430-40f9-ba9a-ba0d1f415f28
DOI: 10.4121/de1df5c0-5430-40f9-ba9a-ba0d1f415f28
Datacite citation style:
He, Gaole; Balayn, Agathe; Buijsman, Stefan; Jie Yang; Gadiraju, Ujwal (2025): Analogy-based explanation evaluation dataset for HCOMP 2022 paper "It Is Like Finding a Polar Bear in the Savannah! Concept-level AI Explanations with Analogical Inference from Commonsense Knowledge.". Version 1. 4TU.ResearchData. software. https://doi.org/10.4121/de1df5c0-5430-40f9-ba9a-ba0d1f415f28.v1
Other citation styles (APA, Harvard, MLA, Vancouver, Chicago, IEEE) available at Datacite
Software
Licence CC BY 4.0
This repo contains all code and data associated with the paper "It Is Like Finding a Polar Bear in the Savannah! Concept-level AI Explanations with Analogical Inference from Commonsense Knowledge." The dataset mainly contains generated analogies and expert evaluations from five colleagues in TU Delft. Our experimental results indicate that the proposed qualitative dimensions can positively contribute to the perceived helpfulness of analogy-based explanations.
History
- 2025-02-06 first online, published, posted
Publisher
4TU.ResearchDataFormat
table/csvAssociated peer-reviewed publication
It Is Like Finding a Polar Bear in the Savannah! Concept-level AI Explanations with Analogical Inference from Commonsense Knowledge.Code hosting project url
https://github.com/delftcrowd/HCOMP2022_ARCHIEOrganizations
TU Delft, Faculty of Electrical Engineering, Mathematics and Computer Science, Department of Software Technology, Web Information Systems GroupDATA
To access the source code, use the following command:
git clone https://data.4tu.nl/v3/datasets/d6208c6f-50b0-42a9-84f5-0da1d39bc805.git "HCOMP2022_ARCHIE"
Files (11)
- 4,485 bytesMD5:
04a98e42091fe266a043951116b45465
README_data_hcomp.md - 16,348 bytesMD5:
abe3e059fe376ed322038b51cd2eb4b7
Analogy_evaluation_expert - E1.csv - 15,909 bytesMD5:
6550ec777a4f6e9a19079e9ad78011c2
Analogy_evaluation_expert - E2.csv - 15,522 bytesMD5:
503a4358e90fbbecb1df80a4d213c83c
Analogy_evaluation_expert - E3.csv - 15,057 bytesMD5:
581325f3c0a195e7c2d33ab14c15ad80
Analogy_evaluation_expert - E4.csv - 16,373 bytesMD5:
bdd1a9a8b226fe8a7b7fd42ad13535ba
Analogy_evaluation_expert - E5.csv - 10,289 bytesMD5:
ce37edad40688e8f8831e73715fdfcd8
calc_Krippendorf_alpha.py - 3,760 bytesMD5:
7ab156964a546030f74b6255c3153da8
calc_spearman_correlation.py - 4,286 bytesMD5:
127fe18ac7a7cbeda12fec84f895b4a9
compare_calorie_place.py - 80,116 bytesMD5:
e515ff5805aacd6b2be6ad721b78e139
generated_analogies.csv - 19,582 bytesMD5:
65e7203328a708e01769dccb39ac5874
statistics_analysis_plot.py -
download all files (zip)
201,727 bytes unzipped