{"artifacts":[{"compression":"none","distribution":"upstream-only","downloads":[{"kind":"upstream","url":"https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/binary/w3a"}],"format":"libsvm","name":"train","sha256":"f050c32d3bd2f18eccbe3847cfe8a3ba9dae4de67d73236013c4a7b9afb48cc7","size":343282},{"compression":"none","distribution":"upstream-only","downloads":[{"kind":"upstream","url":"https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/binary/w3a.t"}],"format":"libsvm","name":"test","sha256":"623e195812435d3c54a0a87120ba9a927ed9771b1d29f9e35fdfa381416b2fb3","size":3134956}],"description":"The w3a binary classification variant of the Web dataset, distributed by LIBSVM with predefined training and test artifacts and 300 sparse features.","license":{"evidence":"The LIBSVM dataset page identifies Platt's JP98a reference as the source but states no license terms for these artifacts: https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/binary.html#w3a. The LIBSVM software license does not establish a license for the distributed dataset.","status":"unknown"},"modality":"tabular","name":"w3a","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#w3a","https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/ref.html#JP98a"],"notes":"LIBSVM documents the feature count and training and test sizes but does not describe feature construction, sample selection, or the split selection procedure for these artifacts.","status":"unknown"}},"provenance":{"authors":["John C. Platt"],"citations":[{"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 this variant 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#w3a","provider":"LIBSVM","retrieved_at":"2026-10-26","upstream_name":"w3a"},"representation":{"decoder":"libsvm-split","decoder_version":1,"expect":{"components":[{"columns":300,"kind":"sparse_matrix","name":"train_features","rows":4912,"type":"float64"},{"kind":"vector","length":4912,"name":"train_response","type":"int64"},{"columns":300,"kind":"sparse_matrix","name":"test_features","rows":44837,"type":"float64"},{"kind":"vector","length":44837,"name":"test_response","type":"int64"}],"verification":[{"algorithm":"sha256","canonical_form":1,"digest":"a8a3ea0cd2552c8afd4f4d5143560d5cae0e6d16f23adfc1686f658f01915581"}]},"inputs":{"test":"test","train":"train"},"options":{"duplicate_features":"error","feature_count":300,"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":"Web (w3a)","version":"1"}