{"meta":{"query_hash":"6061d182248c","filters":{"venue":"Data Science"},"cohort_total":2,"direct_labels_cover":0,"predictions_cover":2,"exported":2,"export_cap":100000,"truncated":false,"label_status":"direct model label, unvalidated","prediction_status":"machine_predicted_unvalidated (Codex and Gemma teacher distillation)","score_status":"score_only:v0-immature-baseline","snapshot":{"source":"OpenAlex, pinned release, all 482 partitions","release":"2026-06-24","frame_built":"2026-07-12"},"permalink":"https://metacan.xera.ac/q/6061d182248c","api":"https://metacan.xera.ac/api/v1/cohort?venue=Data+Science"},"results":[{"id":"W2979856580","doi":"10.3233/ds-190022","title":"String of PURLs – frugal migration and maintenance of persistent identifiers","year":2019,"lang":"en","type":"article","venue":"Data Science","topic":"Scientific Computing and Data Management","field":"Decision Sciences","cited_by":5,"is_retracted":false,"has_abstract":true,"route_ca_aff":true,"route_ca_fund":false,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"CARE Canada","funders":"National Institutes of Health; National Human Genome Research Institute; U.S. Department of Health and Human Services","keywords":"Identifier; Computer science; Key (lock); Software; Computer security; Operating system; Computer network","score_opus":0.15900010087180333,"score_gpt":0.3847611029216242,"score_spread":0.22576100204982086,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W2979856580","genre_codex":"methods","genre_gemma":"methods","domain_codex":null,"domain_gemma":null,"model_version":"metacan-v3-hybrid-931329e0061c","genre_candidate":"methods","genre_consensus":"methods","domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.022093132,0.00052944775,0.850424,0.0044225357,0.0021451432,0.0017664895,0.003883593,0.08088652,0.03384914],"genre_scores_gemma":[0.17445965,0.0006476404,0.704525,0.0054001366,0.0009907862,0.003148544,0.011271054,0.022931188,0.076626025],"study_design_codex":"design_other","study_design_gemma":"not_applicable","domain_scores_codex":[0.97960347,0.0060473806,0.0033648347,0.0038234803,0.0058216746,0.0013390554],"domain_scores_gemma":[0.90779847,0.014537911,0.004580974,0.05815669,0.012738473,0.0021875312],"candidate_categories":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0233088,0.0011347402,0.0011733762,0.005098478,0.0048851366,0.009458324,0.0045000585,0.0031725997,0.015283359],"category_scores_gemma":[0.08769897,0.0016350376,0.0016004355,0.004086187,0.00442352,0.018998599,0.014997973,0.0046939836,0.01240801],"study_design_candidate":"not_applicable","study_design_consensus":null,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_system_candidate":false,"about_ca_system_consensus":false,"study_design_scores_codex":[0.00091378845,0.00045743526,0.014174762,0.0008314253,0.00014038525,0.0009749033,0.0060556075,0.0043597007,0.0104021495,0.25707284,0.20550597,0.49911103],"study_design_scores_gemma":[0.00009861663,0.00019389778,0.0024862457,0.00050061074,0.000072176575,0.0007640052,0.0009427356,0.021548087,0.023558134,0.09508074,0.85449517,0.0002594859],"about_ca_topic_score_codex":0.004650704,"about_ca_topic_score_gemma":0.003242887,"teacher_disagreement_score":0.9905417,"about_ca_system_score_codex":0.0034111168,"about_ca_system_score_gemma":0.0059102336,"threshold_uncertainty_score":0.123270154},"labels":[],"label_agreement":null},{"id":"W4313449255","doi":"10.3233/ds-220057","title":"Towards time-evolving analytics: Online learning for time-dependent evolving data streams","year":2022,"lang":"en","type":"article","venue":"Data Science","topic":"Data Stream Mining Techniques","field":"Computer Science","cited_by":11,"is_retracted":false,"has_abstract":true,"route_ca_aff":true,"route_ca_fund":false,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"Dalhousie University","funders":"","keywords":"Computer science; Analytics; Scalability; Data science; Ambiguity; Data stream mining; Big data; Data analysis; Data mining; Database","score_opus":0.061406812375686806,"score_gpt":0.3311042384424905,"score_spread":0.2696974260668037,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W4313449255","genre_codex":"methods","genre_gemma":"empirical","domain_codex":null,"domain_gemma":null,"model_version":"metacan-v3-hybrid-931329e0061c","genre_candidate":"empirical","genre_consensus":null,"domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.005293958,0.000340092,0.9922312,0.0008321566,0.000046929606,0.000035741887,0.0000909154,0.00052292907,0.00060601707],"genre_scores_gemma":[0.37898737,0.001641311,0.61344016,0.0010013212,0.000579583,0.00029820096,0.0009604871,0.00035128687,0.002740221],"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","domain_scores_codex":[0.99832517,0.0004888715,0.00011134528,0.0005134112,0.00043743785,0.0001236155],"domain_scores_gemma":[0.99030405,0.0066169174,0.00068488717,0.0010923924,0.00091525353,0.00038644456],"candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0044812933,0.0013993804,0.001696022,0.0018938826,0.0007042006,0.0028842129,0.0035480794,0.0018978458,0.001575524],"category_scores_gemma":[0.021313788,0.0007963516,0.0011311366,0.002685664,0.0018408641,0.0074901744,0.0042792326,0.005561076,0.0007243952],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_system_candidate":false,"about_ca_system_consensus":false,"study_design_scores_codex":[0.00019040346,0.00032023867,0.004715739,0.0002964055,0.00017494781,0.00022225514,0.00063931575,0.5575279,0.002377082,0.15243575,0.008744234,0.2723558],"study_design_scores_gemma":[0.000003191781,0.0000099355675,0.000053410415,0.00000749414,0.0000033792494,0.000008360916,0.000016533204,0.9615289,0.00017912485,0.037588816,0.0005967595,0.0000040718132],"about_ca_topic_score_codex":0.004006459,"about_ca_topic_score_gemma":0.0031065722,"teacher_disagreement_score":0.0044812933,"about_ca_system_score_codex":0.0012713592,"about_ca_system_score_gemma":0.0017651216,"threshold_uncertainty_score":0.023699641},"labels":[],"label_agreement":null}]}