<?xml version="1.0" encoding="UTF-8" ?><OAI-PMH xmlns="http://www.openarchives.org/OAI/2.0/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/ http://www.openarchives.org/OAI/2.0/OAI-PMH.xsd"><responseDate>2026-10-14T19:18:31Z</responseDate><request identifier="10.35097/4kjyeg0z2bxmr6eh" metadataPrefix="oai_dc" verb="GetRecord">https://www.radar-service.eu/oai/OAIHandler</request><GetRecord><record><header><identifier>10.35097/4kjyeg0z2bxmr6eh</identifier><datestamp>2026-10-09T10:54:00Z</datestamp><setSpec>radar4kit</setSpec></header><metadata><oai_dc:dc xmlns:dc="http://purl.org/dc/elements/1.1/"
           xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/"
           xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
           xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/ https://www.openarchives.org/OAI/2.0/oai_dc.xsd">
   <dc:identifier>https://dx.doi.org/10.35097/4kjyeg0z2bxmr6eh</dc:identifier>
   <dc:creator>Bach, Jakob</dc:creator>
   <dc:title>Experimental Data for the Dissertation "Leveraging Constraints for User-Centric Feature Selection"</dc:title>
   <dc:publisher>Karlsruhe Institute of Technology</dc:publisher>
   <dc:date>2024</dc:date>
   <dc:subject>Computer Science</dc:subject>
   <dc:type>dataset</dc:type>
   <dc:subject>Dataset</dc:subject>
   <dc:rights>info:eu-repo/semantics/openAccess</dc:rights>
   <dc:rights>https://creativecommons.org/licenses/by/4.0/legalcode</dc:rights>
   <dc:description>These are the experimental data for the dissertation&#xD;
&#xD;
&gt; Bach, Jakob. "Leveraging Constraints for User-Centric Feature Selection"&#xD;
&#xD;
at the [Department of Informatics](https://www.informatik.kit.edu/english/index.php) of the [Karlsruhe Institute of Technology](https://www.kit.edu/english/).&#xD;
You can find the dissertation [here](https://doi.org/10.5445/IR/1000178649).&#xD;
See the `README` for details.&#xD;
&#xD;
Many input datasets (which we also provide here) either&#xD;
&#xD;
- originate from [OpenML](https://www.openml.org) and are CC-BY-licensed or&#xD;
- originate from [PMLB](https://epistasislab.github.io/pmlb/) and are MIT-licensed.&#xD;
&#xD;
Please see the `LICENSE` files in the corresponding `datasets/` subfolders for details.</dc:description>
   <dc:description># Experimental Data for the Dissertation "Leveraging Constraints for User-Centric Feature Selection"&#xD;
&#xD;
These are the experimental data for the dissertation&#xD;
&#xD;
&gt; Bach, Jakob. "Leveraging Constraints for User-Centric Feature Selection"&#xD;
&#xD;
at the [Department of Informatics](https://www.informatik.kit.edu/english/index.php) of the [Karlsruhe Institute of Technology](https://www.kit.edu/english/).&#xD;
&#xD;
The subfolders correspond to individual chapters of the dissertation:&#xD;
&#xD;
- `chap4-syn`: Chapter 4 - "Evaluating the Impact of Constraints on Feature-Selection Results"&#xD;
- `chap5-ms`: Chapter 5 - "Formulating Scientific Hypotheses as Constraints - A Case Study"&#xD;
- `chap6-afs`: Chapter 6 - "Finding Alternative Feature Sets"&#xD;
- `chap7-csd`: Chapter 7 - "Discovering Sparse and Alternative Subgroup Descriptions"&#xD;
&#xD;
See the corresponding `README` files in the subfolders for more information.&#xD;
We already published prior versions of the experimental data, as the dissertation bases on prior papers:&#xD;
&#xD;
- Chapters 4 and 5: [Data](https://doi.org/10.35097/1345) for the [paper](https://doi.org/10.1007/s42979-022-01338-z) "An Empirical Evaluation of Constrained Feature Selection"&#xD;
- Chapter 6: [Data](https://doi.org/10.35097/1920) for the [paper](https://doi.org/10.48550/arXiv.2307.11607) "Finding Optimal Diverse Feature Sets with Alternative Feature Selection" (Version 2)&#xD;
- Chapter 7: [Data](https://doi.org/10.35097/caKKJCtoKqgxyvqG) for the [paper](https://doi.org/10.48550/arXiv.2406.01411) "Using Constraints to Discover Sparse and Alternative Subgroup Descriptions" (Version 1)&#xD;
&#xD;
For Chapters 4, 5, and 7, we mainly consolidate the existing data.&#xD;
In particular, all `*.csv` files (datasets and results) remain unchanged compared to the data linked above.&#xD;
For Chapter 6, we reran the experimental pipeline to integrate a change for the feature-selection method "Greedy Wrapper".&#xD;
The other feature-selection methods have not changed, but experimental data may slightly differ regarding runtimes and for results affected by solver timeouts.&#xD;
&#xD;
For all four chapters, the following files (in each subfolder) differ from prior versions:&#xD;
&#xD;
- `Evaluation_console_output.txt`: The dissertation's evaluation partly differs from the papers' evaluations (e.g., some analyses added, adapted, or removed).&#xD;
- `README.md`: We adapted these files to the context of the dissertation, added some explanations, and proofread them.</dc:description>
   <dc:subject>feature selection</dc:subject>
   <dc:subject>subgroup discovery</dc:subject>
   <dc:subject>constraints</dc:subject>
   <dc:subject>alternatives</dc:subject>
   <dc:subject>explainability</dc:subject>
   <dc:subject>interpretability</dc:subject>
   <dc:subject>XAI</dc:subject>
   <dc:relation>https://publikationen.bibliothek.kit.edu/1000175819</dc:relation>
   <dc:format>application/x-tar</dc:format>
</oai_dc:dc></metadata></record></GetRecord></OAI-PMH>