Changes
On November 28, 2024 at 1:16:49 PM UTC, admin:
-
Changed value of field
extra_authors
to[{'extra_author': 'Zoller, Kolja', 'familyName': 'Zoller', 'givenName': 'Kolja', 'orcid': ''}, {'extra_author': 'Schulz, Katrin', 'familyName': 'Schulz', 'givenName': 'Katrin', 'orcid': ''}]
in Experimental data for the paper ''an empirical evaluation of constrained feature selection"
f | 1 | { | f | 1 | { |
2 | "author": "Bach, Jakob", | 2 | "author": "Bach, Jakob", | ||
3 | "author_email": "", | 3 | "author_email": "", | ||
n | n | 4 | "citation": [], | ||
4 | "creator_user_id": "17755db4-395a-4b3b-ac09-e8e3484ca700", | 5 | "creator_user_id": "17755db4-395a-4b3b-ac09-e8e3484ca700", | ||
5 | "doi": "10.35097/1345", | 6 | "doi": "10.35097/1345", | ||
6 | "doi_date_published": "2023", | 7 | "doi_date_published": "2023", | ||
7 | "doi_publisher": "", | 8 | "doi_publisher": "", | ||
8 | "doi_status": "True", | 9 | "doi_status": "True", | ||
9 | "extra_authors": [ | 10 | "extra_authors": [ | ||
10 | { | 11 | { | ||
11 | "extra_author": "Zoller, Kolja", | 12 | "extra_author": "Zoller, Kolja", | ||
n | n | 13 | "familyName": "Zoller", | ||
14 | "givenName": "Kolja", | ||||
12 | "orcid": "" | 15 | "orcid": "" | ||
13 | }, | 16 | }, | ||
14 | { | 17 | { | ||
15 | "extra_author": "Schulz, Katrin", | 18 | "extra_author": "Schulz, Katrin", | ||
n | n | 19 | "familyName": "Schulz", | ||
20 | "givenName": "Katrin", | ||||
16 | "orcid": "" | 21 | "orcid": "" | ||
17 | } | 22 | } | ||
18 | ], | 23 | ], | ||
n | n | 24 | "familyName": "Bach", | ||
25 | "givenName": "Jakob", | ||||
19 | "groups": [], | 26 | "groups": [], | ||
20 | "id": "8a78f8b2-4c06-4c34-aa60-f1d9d3ce8471", | 27 | "id": "8a78f8b2-4c06-4c34-aa60-f1d9d3ce8471", | ||
21 | "isopen": false, | 28 | "isopen": false, | ||
22 | "license_id": "CC BY 4.0 Attribution", | 29 | "license_id": "CC BY 4.0 Attribution", | ||
23 | "license_title": "CC BY 4.0 Attribution", | 30 | "license_title": "CC BY 4.0 Attribution", | ||
24 | "metadata_created": "2023-08-04T08:50:34.911600", | 31 | "metadata_created": "2023-08-04T08:50:34.911600", | ||
n | 25 | "metadata_modified": "2023-08-04T09:29:05.791207", | n | 32 | "metadata_modified": "2024-11-28T13:16:49.228983", |
26 | "name": "rdr-doi-10-35097-1345", | 33 | "name": "rdr-doi-10-35097-1345", | ||
27 | "notes": "Abstract: These are the experimental data for the | 34 | "notes": "Abstract: These are the experimental data for the | ||
28 | paper\r\n\r\n> Bach, Jakob, et al. \"An Empirical Evaluation of | 35 | paper\r\n\r\n> Bach, Jakob, et al. \"An Empirical Evaluation of | ||
29 | Constrained Feature Selection\"\r\n\r\npublished at the journal [*SN | 36 | Constrained Feature Selection\"\r\n\r\npublished at the journal [*SN | ||
30 | Computer Science*](https://www.springer.com/journal/42979).\r\nYou can | 37 | Computer Science*](https://www.springer.com/journal/42979).\r\nYou can | ||
31 | find the paper [here](https://doi.org/10.1007/s42979-022-01338-z) and | 38 | find the paper [here](https://doi.org/10.1007/s42979-022-01338-z) and | ||
32 | the code | 39 | the code | ||
33 | ://github.com/Jakob-Bach/Constrained-Filter-Feature-Selection).\r\nSee | 40 | ://github.com/Jakob-Bach/Constrained-Filter-Feature-Selection).\r\nSee | ||
34 | the `README` for details.\r\n\r\nSome of the datasets used in our | 41 | the `README` for details.\r\n\r\nSome of the datasets used in our | ||
35 | study (which we also provide here) originate from | 42 | study (which we also provide here) originate from | ||
36 | [OpenML](https://www.openml.org) and are CC-BY-licensed.\r\nPlease see | 43 | [OpenML](https://www.openml.org) and are CC-BY-licensed.\r\nPlease see | ||
37 | the paragraph `Licensing` in the `README` for details, e.g., on the | 44 | the paragraph `Licensing` in the `README` for details, e.g., on the | ||
38 | authors of these datasets.\r\nTechnicalRemarks: # Experimental Data | 45 | authors of these datasets.\r\nTechnicalRemarks: # Experimental Data | ||
39 | for the Paper \"An Empirical Evaluation of Constrained Feature | 46 | for the Paper \"An Empirical Evaluation of Constrained Feature | ||
40 | Selection\"\r\n\r\nThese are the experimental data for the | 47 | Selection\"\r\n\r\nThese are the experimental data for the | ||
41 | paper\r\n\r\n> Bach, Jakob, et al. \"An Empirical Evaluation of | 48 | paper\r\n\r\n> Bach, Jakob, et al. \"An Empirical Evaluation of | ||
42 | Constrained Feature Selection\"\r\n\r\naccepted at the journal [*SN | 49 | Constrained Feature Selection\"\r\n\r\naccepted at the journal [*SN | ||
43 | Computer | 50 | Computer | ||
44 | Science*](https://www.springer.com/journal/42979).\r\n\r\nCheck our | 51 | Science*](https://www.springer.com/journal/42979).\r\n\r\nCheck our | ||
45 | [GitHub | 52 | [GitHub | ||
46 | y](https://github.com/Jakob-Bach/Constrained-Filter-Feature-Selection) | 53 | y](https://github.com/Jakob-Bach/Constrained-Filter-Feature-Selection) | ||
47 | for the code and instructions to reproduce the experiments.\r\nThe | 54 | for the code and instructions to reproduce the experiments.\r\nThe | ||
48 | data were obtained on a server with an `AMD EPYC 7551` | 55 | data were obtained on a server with an `AMD EPYC 7551` | ||
49 | [CPU](https://www.amd.com/en/products/cpu/amd-epyc-7551) (32 physical | 56 | [CPU](https://www.amd.com/en/products/cpu/amd-epyc-7551) (32 physical | ||
50 | cores, base clock of 2.0 GHz) and 128 GB RAM.\r\nThe Python version | 57 | cores, base clock of 2.0 GHz) and 128 GB RAM.\r\nThe Python version | ||
51 | was `3.8`.\r\nOur paper contains two studies, and we provide data for | 58 | was `3.8`.\r\nOur paper contains two studies, and we provide data for | ||
52 | both of them.\r\n\r\nRunning the experimental pipeline for the study | 59 | both of them.\r\n\r\nRunning the experimental pipeline for the study | ||
53 | with synthetic constraints (`syn_pipeline.py`) took several | 60 | with synthetic constraints (`syn_pipeline.py`) took several | ||
54 | hours.\r\nThe commit hash for the last run of this pipeline is | 61 | hours.\r\nThe commit hash for the last run of this pipeline is | ||
55 | ature-Selection/tree/acc34cf5d22b0a8427852a01288bb8b34f5d8c98).\r\nThe | 62 | ature-Selection/tree/acc34cf5d22b0a8427852a01288bb8b34f5d8c98).\r\nThe | ||
56 | commit hash for the last run of the corresponding evaluation | 63 | commit hash for the last run of the corresponding evaluation | ||
57 | (`syn_evaluation.py`) is | 64 | (`syn_evaluation.py`) is | ||
58 | lection/tree/c1a7e7e99e56c1a178a602596c13641d7771df0a).\r\n\r\nRunning | 65 | lection/tree/c1a7e7e99e56c1a178a602596c13641d7771df0a).\r\n\r\nRunning | ||
59 | the experimental pipeline for the case study in materials science | 66 | the experimental pipeline for the case study in materials science | ||
60 | (`ms_pipeline.py`) took less than one hour.\r\nThe commit hash for the | 67 | (`ms_pipeline.py`) took less than one hour.\r\nThe commit hash for the | ||
61 | last run of this pipeline is | 68 | last run of this pipeline is | ||
62 | ature-Selection/tree/ba30bf9f11703e2a8a942425e2cd4b9f36ead513).\r\nThe | 69 | ature-Selection/tree/ba30bf9f11703e2a8a942425e2cd4b9f36ead513).\r\nThe | ||
63 | commit hash for the last run of the corresponding evaluation | 70 | commit hash for the last run of the corresponding evaluation | ||
64 | (`ms_evaluation.py`) is | 71 | (`ms_evaluation.py`) is | ||
65 | e-Selection/tree/c1a7e7e99e56c1a178a602596c13641d7771df0a).\r\n\r\nAll | 72 | e-Selection/tree/c1a7e7e99e56c1a178a602596c13641d7771df0a).\r\n\r\nAll | ||
66 | these commits are also tagged.\r\n\r\nIn the following, we describe | 73 | these commits are also tagged.\r\n\r\nIn the following, we describe | ||
67 | the structure/content of each data file.\r\nAll files are plain CSVs, | 74 | the structure/content of each data file.\r\nAll files are plain CSVs, | ||
68 | so you can read them with `pandas.read_csv()`.\r\n\r\n## | 75 | so you can read them with `pandas.read_csv()`.\r\n\r\n## | ||
69 | `ms/`\r\n\r\nThe input data for the case study in materials science | 76 | `ms/`\r\n\r\nThe input data for the case study in materials science | ||
70 | (`ms_pipeline.py`).\r\nOutput of the script | 77 | (`ms_pipeline.py`).\r\nOutput of the script | ||
71 | `prepare_ms_dataset.py`.\r\nAs the raw simulation dataset is quite | 78 | `prepare_ms_dataset.py`.\r\nAs the raw simulation dataset is quite | ||
72 | large, we only provide a pre-processed version of it\r\n(we do not | 79 | large, we only provide a pre-processed version of it\r\n(we do not | ||
73 | provide the input to `prepare_ms_dataset.py`).\r\nIn this | 80 | provide the input to `prepare_ms_dataset.py`).\r\nIn this | ||
74 | pre-processed version, the feature and target parts of the data are | 81 | pre-processed version, the feature and target parts of the data are | ||
75 | already separated into two files: `voxel_data_predict_glissile_X.csv` | 82 | already separated into two files: `voxel_data_predict_glissile_X.csv` | ||
76 | and `voxel_data_predict_glissile_y.csv`.\r\nIn | 83 | and `voxel_data_predict_glissile_y.csv`.\r\nIn | ||
77 | `voxel_data_predict_glissile_X.csv`, each column is a numeric | 84 | `voxel_data_predict_glissile_X.csv`, each column is a numeric | ||
78 | feature.\r\n`voxel_data_predict_glissile_y.csv` only contains one | 85 | feature.\r\n`voxel_data_predict_glissile_y.csv` only contains one | ||
79 | column, the numeric prediction target (reaction density of glissile | 86 | column, the numeric prediction target (reaction density of glissile | ||
80 | reactions).\r\n\r\n## `ms-results/`\r\n\r\nOnly contains one result | 87 | reactions).\r\n\r\n## `ms-results/`\r\n\r\nOnly contains one result | ||
81 | file (`results.csv`) for the case study in materials | 88 | file (`results.csv`) for the case study in materials | ||
82 | science.\r\nOutput of the script `ms_pipeline.py`, input to the script | 89 | science.\r\nOutput of the script `ms_pipeline.py`, input to the script | ||
83 | `ms_evaluation.py`.\r\nThe columns of the file mostly correspond to | 90 | `ms_evaluation.py`.\r\nThe columns of the file mostly correspond to | ||
84 | evaluation metrics used in the paper;\r\nsee Appendix A.1 for | 91 | evaluation metrics used in the paper;\r\nsee Appendix A.1 for | ||
85 | definitions of the latter.\r\n\r\n- `objective_value` (float): | 92 | definitions of the latter.\r\n\r\n- `objective_value` (float): | ||
86 | Objective `Q(s, X, y)`, the sum of the qualities of the selected | 93 | Objective `Q(s, X, y)`, the sum of the qualities of the selected | ||
87 | features.\r\n- `num_selected` (int): `n_{se}`, the number of selected | 94 | features.\r\n- `num_selected` (int): `n_{se}`, the number of selected | ||
88 | features.\r\n- `selected` (string, but actually a list of strings): | 95 | features.\r\n- `selected` (string, but actually a list of strings): | ||
89 | Names of the selected features.\r\n- `num_variables` (int): `n`, the | 96 | Names of the selected features.\r\n- `num_variables` (int): `n`, the | ||
90 | total number of features in the dataset.\r\n- | 97 | total number of features in the dataset.\r\n- | ||
91 | `num_constrained_variables` (int): `n_{cf}`, the number of features | 98 | `num_constrained_variables` (int): `n_{cf}`, the number of features | ||
92 | involved in constraints.\r\n- `num_unique_constrained_variables` | 99 | involved in constraints.\r\n- `num_unique_constrained_variables` | ||
93 | (int): `n_{ucf}`, the number of unique features involved in | 100 | (int): `n_{ucf}`, the number of unique features involved in | ||
94 | constraints.\r\n- `num_constraints` (int): `n_{co}`, the number of | 101 | constraints.\r\n- `num_constraints` (int): `n_{co}`, the number of | ||
95 | constraints.\r\n- `frac_solutions` (float): `n_{so}^{norm}`, the | 102 | constraints.\r\n- `frac_solutions` (float): `n_{so}^{norm}`, the | ||
96 | number of valid (regarding constraints) feature sets relative to the | 103 | number of valid (regarding constraints) feature sets relative to the | ||
97 | total number of feature sets.\r\n- `linear-regression_train_r2` | 104 | total number of feature sets.\r\n- `linear-regression_train_r2` | ||
98 | (float): `R^2` (coefficient of determination) for linear-regression | 105 | (float): `R^2` (coefficient of determination) for linear-regression | ||
99 | models, trained with the selected features, predicting on the training | 106 | models, trained with the selected features, predicting on the training | ||
100 | set.\r\n- `linear-regression_test_r2` (float): `R^2` (coefficient of | 107 | set.\r\n- `linear-regression_test_r2` (float): `R^2` (coefficient of | ||
101 | determination) for linear-regression models, trained with the selected | 108 | determination) for linear-regression models, trained with the selected | ||
102 | features, predicting on the test set.\r\n- `regression-tree_train_r2` | 109 | features, predicting on the test set.\r\n- `regression-tree_train_r2` | ||
103 | (float): `R^2` (coefficient of determination) for regression-tree | 110 | (float): `R^2` (coefficient of determination) for regression-tree | ||
104 | models, trained with the selected features, predicting on the training | 111 | models, trained with the selected features, predicting on the training | ||
105 | set.\r\n- `regression-tree_test_r2` (float): `R^2` (coefficient of | 112 | set.\r\n- `regression-tree_test_r2` (float): `R^2` (coefficient of | ||
106 | determination) for regression-tree models, trained with the selected | 113 | determination) for regression-tree models, trained with the selected | ||
107 | features, predicting on the test set.\r\n- `xgb-linear_train_r2` | 114 | features, predicting on the test set.\r\n- `xgb-linear_train_r2` | ||
108 | (float): `R^2` (coefficient of determination) for linear XGBoost | 115 | (float): `R^2` (coefficient of determination) for linear XGBoost | ||
109 | models, trained with the selected features, predicting on the training | 116 | models, trained with the selected features, predicting on the training | ||
110 | set.\r\n- `xgb-linear_test_r2` (float): `R^2` (coefficient of | 117 | set.\r\n- `xgb-linear_test_r2` (float): `R^2` (coefficient of | ||
111 | determination) for linear XGBoost models, trained with the selected | 118 | determination) for linear XGBoost models, trained with the selected | ||
112 | features, predicting on the test set.\r\n- `xgb-tree_train_r2` | 119 | features, predicting on the test set.\r\n- `xgb-tree_train_r2` | ||
113 | (float): `R^2` (coefficient of determination) for tree-based XGBoost | 120 | (float): `R^2` (coefficient of determination) for tree-based XGBoost | ||
114 | models, trained with the selected features, predicting on the training | 121 | models, trained with the selected features, predicting on the training | ||
115 | set.\r\n- `xgb-tree_test_r2` (float): `R^2` (coefficient of | 122 | set.\r\n- `xgb-tree_test_r2` (float): `R^2` (coefficient of | ||
116 | determination) for tree-based XGBoost models, trained with the | 123 | determination) for tree-based XGBoost models, trained with the | ||
117 | selected features, predicting on the test set.\r\n- `evaluation_time` | 124 | selected features, predicting on the test set.\r\n- `evaluation_time` | ||
118 | (float): Runtime (in s) for evaluating one set of constraints.\r\n- | 125 | (float): Runtime (in s) for evaluating one set of constraints.\r\n- | ||
119 | `split_idx` (int): Index of the cross-validation fold.\r\n- | 126 | `split_idx` (int): Index of the cross-validation fold.\r\n- | ||
120 | `quality_name` (string): Measure for feature quality (absolute | 127 | `quality_name` (string): Measure for feature quality (absolute | ||
121 | correlation or mutual information).\r\n- `constraint_name` (string): | 128 | correlation or mutual information).\r\n- `constraint_name` (string): | ||
122 | Name of the constraint type (see paper).\r\n- `dataset_name` (string): | 129 | Name of the constraint type (see paper).\r\n- `dataset_name` (string): | ||
123 | Name of the dataset.\r\n\r\n## `openml/`\r\n\r\nThe input data for the | 130 | Name of the dataset.\r\n\r\n## `openml/`\r\n\r\nThe input data for the | ||
124 | study with synthetic constraints (`syn_pipeline.py`).\r\nOutput of the | 131 | study with synthetic constraints (`syn_pipeline.py`).\r\nOutput of the | ||
125 | script `prepare_openml_datasets.py`.\r\nWe downloaded 35 datasets from | 132 | script `prepare_openml_datasets.py`.\r\nWe downloaded 35 datasets from | ||
126 | [OpenML](https://www.openml.org) and removed non-numeric | 133 | [OpenML](https://www.openml.org) and removed non-numeric | ||
127 | columns.\r\nAlso, we separated the feature part (`*_X.csv`) and the | 134 | columns.\r\nAlso, we separated the feature part (`*_X.csv`) and the | ||
128 | target part (`*_y.csv`) of each dataset.\r\n`_data_overview.csv` | 135 | target part (`*_y.csv`) of each dataset.\r\n`_data_overview.csv` | ||
129 | contains meta-data for the datasets, including dataset id, dataset | 136 | contains meta-data for the datasets, including dataset id, dataset | ||
130 | version, and uploader.\r\n\r\n**Licensing**\r\n\r\nPlease consult each | 137 | version, and uploader.\r\n\r\n**Licensing**\r\n\r\nPlease consult each | ||
131 | dataset's website on [OpenML](https://www.openml.org) for licensing | 138 | dataset's website on [OpenML](https://www.openml.org) for licensing | ||
132 | information and citation requests.\r\nAccording to OpenML's | 139 | information and citation requests.\r\nAccording to OpenML's | ||
133 | [terms](https://www.openml.org/terms), OpenML datasets fall under the | 140 | [terms](https://www.openml.org/terms), OpenML datasets fall under the | ||
134 | [CC-BY](https://creativecommons.org/licenses/by/4.0/) license.\r\nThe | 141 | [CC-BY](https://creativecommons.org/licenses/by/4.0/) license.\r\nThe | ||
135 | datasets used in our study were uploaded by:\r\n\r\n- Jan van Rijn | 142 | datasets used in our study were uploaded by:\r\n\r\n- Jan van Rijn | ||
136 | (user id: 1)\r\n- Joaquin Vanschoren (user id: 2)\r\n- Rafael Gomes | 143 | (user id: 1)\r\n- Joaquin Vanschoren (user id: 2)\r\n- Rafael Gomes | ||
137 | Mantovani (user id: 64)\r\n- Tobias Kuehn (user id: 94)\r\n- Richard | 144 | Mantovani (user id: 64)\r\n- Tobias Kuehn (user id: 94)\r\n- Richard | ||
138 | Ooms (user id: 8684)\r\n- R P (user id: 15317)\r\n\r\nSee | 145 | Ooms (user id: 8684)\r\n- R P (user id: 15317)\r\n\r\nSee | ||
139 | `_data_overview.csv` to match each dataset to its uploader.\r\n\r\n## | 146 | `_data_overview.csv` to match each dataset to its uploader.\r\n\r\n## | ||
140 | `openml-results/`\r\n\r\nResult files for the study with synthetic | 147 | `openml-results/`\r\n\r\nResult files for the study with synthetic | ||
141 | constraints.\r\nOutput of the script `syn_pipeline.py`, input to the | 148 | constraints.\r\nOutput of the script `syn_pipeline.py`, input to the | ||
142 | script `syn_evaluation.py`.\r\nOne result file for each combination of | 149 | script `syn_evaluation.py`.\r\nOne result file for each combination of | ||
143 | the 10 constraint generators and the 35 datasets, plus one overall | 150 | the 10 constraint generators and the 35 datasets, plus one overall | ||
144 | (merged) file, `results.csv`.\r\nThe columns of the result files are | 151 | (merged) file, `results.csv`.\r\nThe columns of the result files are | ||
145 | the those of `ms-results/results.csv`, minus `selected` and | 152 | the those of `ms-results/results.csv`, minus `selected` and | ||
146 | `evaluation_time`;\r\nsee above for detailed descriptions.", | 153 | `evaluation_time`;\r\nsee above for detailed descriptions.", | ||
147 | "num_resources": 0, | 154 | "num_resources": 0, | ||
148 | "num_tags": 4, | 155 | "num_tags": 4, | ||
149 | "orcid": "0000-0003-0301-2798", | 156 | "orcid": "0000-0003-0301-2798", | ||
150 | "organization": { | 157 | "organization": { | ||
151 | "approval_status": "approved", | 158 | "approval_status": "approved", | ||
152 | "created": "2023-01-12T13:30:23.238233", | 159 | "created": "2023-01-12T13:30:23.238233", | ||
153 | "description": "RADAR (Research Data Repository) is a | 160 | "description": "RADAR (Research Data Repository) is a | ||
154 | cross-disciplinary repository for archiving and publishing research | 161 | cross-disciplinary repository for archiving and publishing research | ||
155 | data from completed scientific studies and projects. The focus is on | 162 | data from completed scientific studies and projects. The focus is on | ||
156 | research data from subjects that do not yet have their own | 163 | research data from subjects that do not yet have their own | ||
157 | discipline-specific infrastructures for research data management. ", | 164 | discipline-specific infrastructures for research data management. ", | ||
158 | "id": "013c89a9-383c-4200-8baa-0f78bf1d91f9", | 165 | "id": "013c89a9-383c-4200-8baa-0f78bf1d91f9", | ||
159 | "image_url": "radar-logo.svg", | 166 | "image_url": "radar-logo.svg", | ||
160 | "is_organization": true, | 167 | "is_organization": true, | ||
161 | "name": "radar", | 168 | "name": "radar", | ||
162 | "state": "active", | 169 | "state": "active", | ||
163 | "title": "RADAR", | 170 | "title": "RADAR", | ||
164 | "type": "organization" | 171 | "type": "organization" | ||
165 | }, | 172 | }, | ||
166 | "owner_org": "013c89a9-383c-4200-8baa-0f78bf1d91f9", | 173 | "owner_org": "013c89a9-383c-4200-8baa-0f78bf1d91f9", | ||
167 | "private": false, | 174 | "private": false, | ||
168 | "production_year": "2021", | 175 | "production_year": "2021", | ||
169 | "publication_year": "2023", | 176 | "publication_year": "2023", | ||
170 | "publishers": [ | 177 | "publishers": [ | ||
171 | { | 178 | { | ||
172 | "publisher": "Karlsruhe Institute of Technology" | 179 | "publisher": "Karlsruhe Institute of Technology" | ||
173 | } | 180 | } | ||
174 | ], | 181 | ], | ||
t | t | 182 | "related_identifiers": [ | ||
183 | { | ||||
184 | "identifier": | ||||
185 | "https://publikationen.bibliothek.kit.edu/1000148891", | ||||
186 | "identifier_type": "URL", | ||||
187 | "relation_type": "IsIdenticalTo" | ||||
188 | } | ||||
189 | ], | ||||
175 | "relationships_as_object": [], | 190 | "relationships_as_object": [], | ||
176 | "relationships_as_subject": [], | 191 | "relationships_as_subject": [], | ||
177 | "repository_name": "RADAR (Research Data Repository)", | 192 | "repository_name": "RADAR (Research Data Repository)", | ||
178 | "resources": [], | 193 | "resources": [], | ||
179 | "services_used_list": "", | 194 | "services_used_list": "", | ||
180 | "source_metadata_created": "2023", | 195 | "source_metadata_created": "2023", | ||
181 | "source_metadata_modified": "", | 196 | "source_metadata_modified": "", | ||
182 | "state": "active", | 197 | "state": "active", | ||
183 | "subject_areas": [ | 198 | "subject_areas": [ | ||
184 | { | 199 | { | ||
185 | "subject_area_additional": "", | 200 | "subject_area_additional": "", | ||
186 | "subject_area_name": "Computer Science" | 201 | "subject_area_name": "Computer Science" | ||
187 | } | 202 | } | ||
188 | ], | 203 | ], | ||
189 | "tags": [ | 204 | "tags": [ | ||
190 | { | 205 | { | ||
191 | "display_name": "Constraints", | 206 | "display_name": "Constraints", | ||
192 | "id": "fa3d947e-eff8-42bd-953b-4f77336f1a66", | 207 | "id": "fa3d947e-eff8-42bd-953b-4f77336f1a66", | ||
193 | "name": "Constraints", | 208 | "name": "Constraints", | ||
194 | "state": "active", | 209 | "state": "active", | ||
195 | "vocabulary_id": null | 210 | "vocabulary_id": null | ||
196 | }, | 211 | }, | ||
197 | { | 212 | { | ||
198 | "display_name": "Domain knowledge", | 213 | "display_name": "Domain knowledge", | ||
199 | "id": "6f0a4c92-efd1-44a7-ba68-522c71b504c5", | 214 | "id": "6f0a4c92-efd1-44a7-ba68-522c71b504c5", | ||
200 | "name": "Domain knowledge", | 215 | "name": "Domain knowledge", | ||
201 | "state": "active", | 216 | "state": "active", | ||
202 | "vocabulary_id": null | 217 | "vocabulary_id": null | ||
203 | }, | 218 | }, | ||
204 | { | 219 | { | ||
205 | "display_name": "Feature selection", | 220 | "display_name": "Feature selection", | ||
206 | "id": "29a81ebe-698a-43e0-97bb-b4361d1f9d29", | 221 | "id": "29a81ebe-698a-43e0-97bb-b4361d1f9d29", | ||
207 | "name": "Feature selection", | 222 | "name": "Feature selection", | ||
208 | "state": "active", | 223 | "state": "active", | ||
209 | "vocabulary_id": null | 224 | "vocabulary_id": null | ||
210 | }, | 225 | }, | ||
211 | { | 226 | { | ||
212 | "display_name": "Theory-guided data science", | 227 | "display_name": "Theory-guided data science", | ||
213 | "id": "0ee4c6b8-29e1-4304-9112-2aef37a5b8b8", | 228 | "id": "0ee4c6b8-29e1-4304-9112-2aef37a5b8b8", | ||
214 | "name": "Theory-guided data science", | 229 | "name": "Theory-guided data science", | ||
215 | "state": "active", | 230 | "state": "active", | ||
216 | "vocabulary_id": null | 231 | "vocabulary_id": null | ||
217 | } | 232 | } | ||
218 | ], | 233 | ], | ||
219 | "title": "Experimental data for the paper ''an empirical evaluation | 234 | "title": "Experimental data for the paper ''an empirical evaluation | ||
220 | of constrained feature selection\"", | 235 | of constrained feature selection\"", | ||
221 | "type": "vdataset", | 236 | "type": "vdataset", | ||
222 | "url": "https://doi.org/10.35097/1345" | 237 | "url": "https://doi.org/10.35097/1345" | ||
223 | } | 238 | } |