{"artifacts":[{"compression":"none","distribution":"upstream-only","downloads":[{"kind":"upstream","url":"https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/binary/a5a"}],"format":"libsvm","name":"train","sha256":"5cb3a6603de0a0eb43afcdd45b1820c71ee7f20da1c09f054696bb71cbc45ad8","size":458892},{"compression":"none","distribution":"upstream-only","downloads":[{"kind":"upstream","url":"https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/binary/a5a.t"}],"format":"libsvm","name":"test","sha256":"9ddbc60c0a3c5a8249e51f663d7110c4050e40d67301640418e9f44186db1bd1","size":1870983}],"description":"The a5a binary classification variant of the UCI Adult census income dataset, distributed by LIBSVM with predefined training and test artifacts. The task predicts whether annual income exceeds $50,000 using 123 binary features derived from the original 14 attributes.","license":{"evidence":"UCI identifies Barry Becker and Ronny Kohavi as the Adult dataset creators and licenses the original dataset under CC BY 4.0: https://archive.ics.uci.edu/dataset/2/adult. The LIBSVM page identifies Adult as the source and documents its feature encoding but states no license terms for the transformed artifacts: https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/binary.html#a5a. The applicability of the original license to these distributed artifacts has not been established.","status":"unknown"},"modality":"tabular","name":"a5a","preprocessing":{"datamonger":{"evidence":["spec/decoders/libsvm-v1.md"],"notes":"Datamonger decodes the provider artifacts without additional statistical preprocessing.","status":"none"},"upstream":{"evidence":["https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/binary.html#a5a","https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/binary.html#a1a"],"notes":"LIBSVM documents 123 features and the training and test sizes. The quantile boundaries, fitting population, handling of original missing values, and split selection procedure have not been established.","status":"documented","steps":[{"description":"The six continuous Adult attributes are discretized into quantiles and represented by binary indicators. Each category of the eight categorical attributes is represented by a binary indicator.","kind":"feature-encoding"}]}},"provenance":{"authors":["Barry Becker","Ronny Kohavi"],"citations":[{"citation":"Becker, B., and Kohavi, R. (1996). Adult [Dataset]. UCI Machine Learning Repository. https://doi.org/10.24432/C5XW20.","note":"Original Adult dataset, as cited by UCI.","role":"dataset"},{"citation":"John C. Platt. Fast training of support vector machines using sequential minimal optimization. In Advances in Kernel Methods - Support Vector Learning. MIT Press, 1998.","note":"Reference JP98a associated with these variants by LIBSVM: https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/ref.html#JP98a.","role":"paper"}],"landing_page":"https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/binary.html#a5a","provider":"LIBSVM","retrieved_at":"2026-10-26","upstream_name":"a5a"},"representation":{"decoder":"libsvm-split","decoder_version":1,"expect":{"components":[{"columns":123,"kind":"sparse_matrix","name":"train_features","rows":6414,"type":"float64"},{"kind":"vector","length":6414,"name":"train_response","type":"int64"},{"columns":123,"kind":"sparse_matrix","name":"test_features","rows":26147,"type":"float64"},{"kind":"vector","length":26147,"name":"test_response","type":"int64"}],"verification":[{"algorithm":"sha256","canonical_form":1,"digest":"95f3880f8952bb14247ca30f33c7c45daeefc18646b0a737cb55c585f29bd567"}]},"inputs":{"test":"test","train":"train"},"options":{"duplicate_features":"error","feature_count":123,"index_base":1,"label_type":"int64","row_order":"source","target_name":"response"}},"schema_version":4,"source":"libsvm","tasks":[{"name":"default","splits":{"test":{"features":"test_features","target":"test_response"},"train":{"features":"train_features","target":"train_response"}},"type":"classification"}],"title":"Adult (a5a)","version":"1"}