{"meta":{"page":1,"per_page":50,"max_per_page":100,"total":2,"total_is_capped":false,"direct_labels_cover":0,"predictions_cover":2,"direct_label_status":"direct model label, unvalidated","prediction_status":"machine_predicted_unvalidated (Codex and Gemma teacher distillation)","score_status":"score_only:v0-immature-baseline (scores rank; they never assert a category)","snapshot":{"source":"OpenAlex, pinned release, all 482 partitions","release":"2026-06-24","frame_built":"2026-07-12"},"query_hash":"e2f6e0948738","filters":{"venue":"Australasian Data Mining Conference"}},"results":[{"id":"W2162987981","doi":"","title":"A characterization of wordnet features in Boolean models for text classification","year":2006,"lang":"en","type":"article","venue":"Australasian Data Mining Conference","topic":"Text and Document Classification Technologies","field":"Computer Science","cited_by":18,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Regina","funders":"","keywords":"WordNet; Computer science; Artificial intelligence; Naive Bayes classifier; Natural language processing; Classifier (UML); Support vector machine; Weighting; Machine learning; Information retrieval","retraction":null,"screen_n_in":null,"score":{"opus":0.08786756615368328,"gpt":0.2996306411300966,"spread":0.2117630749764133,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002154147,0.0001302377,0.0001715171,0.0002176839,0.000046295,0.0001668876,0.001754175,0.0001105316,0.000006645283],"category_scores_gemma":[0.00005827891,0.0001320119,0.00002103956,0.0003761001,0.0000806949,0.001581641,0.0002332567,0.00008989969,0.000003281437],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002400158,"about_ca_system_score_gemma":0.00009273928,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000462287,"about_ca_topic_score_gemma":0.0001002785,"domain_scores_codex":[0.9987105,0.00002694645,0.000379669,0.0005035747,0.0001575186,0.0002218585],"domain_scores_gemma":[0.9983733,0.000061697,0.0002726864,0.001184184,0.00008325969,0.00002488141],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"observational","study_design_scores_codex":[0.00001969556,0.0001569599,0.005425821,0.00005629938,0.000009771071,0.000001589254,0.0005671293,0.00003654913,0.09066816,0.6431443,0.002416969,0.2574967],"study_design_scores_gemma":[0.00127153,0.0001757101,0.5208303,0.0002714995,0.00002226438,0.000009031654,0.0009065743,0.3923497,0.03126755,0.04726565,0.004946199,0.000683977],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1292161,0.00001832204,0.8656199,0.003059448,0.0001101397,0.0004250691,0.0001636401,0.0002318218,0.001155547],"genre_scores_gemma":[0.9435852,0.00001310358,0.05495611,0.00001858726,0.00002058318,0.00004521989,0.0008172205,0.000006908325,0.0005371289],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.814369,"threshold_uncertainty_score":0.5383291,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2293475867","doi":"","title":"Towards web search engine scale data mining","year":2009,"lang":"en","type":"article","venue":"Australasian Data Mining Conference","topic":"Data Management and Algorithms","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Simon Fraser University","funders":"","keywords":"Computer science; Scalability; Adaptability; Search engine; Web mining; Data science; Data mining; Context (archaeology); Web search query; Web crawler; Web search engine; Metasearch engine; Search analytics; Construct (python library); Big data; Information retrieval; World Wide Web; Database; Web service","retraction":null,"screen_n_in":null,"score":{"opus":0.1295590154397186,"gpt":0.3325322425937539,"spread":0.2029732271540353,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","scholarly_communication","open_science"],"consensus_categories":[],"category_scores_codex":[0.001124394,0.0003345875,0.0003316863,0.0002324619,0.0002026675,0.001397634,0.01519033,0.00009664104,0.0001528119],"category_scores_gemma":[0.000140725,0.0003351947,0.00002897187,0.0007453918,0.0001099339,0.005879886,0.006175507,0.000306537,0.0001799481],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002504665,"about_ca_system_score_gemma":0.0003198091,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004536323,"about_ca_topic_score_gemma":0.00004323543,"domain_scores_codex":[0.996307,0.00008653157,0.0004314037,0.001669084,0.0006730146,0.0008329715],"domain_scores_gemma":[0.9922827,0.00007112055,0.0001166341,0.007171733,0.00008556889,0.0002722383],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000004251974,0.00008069877,0.0006312301,0.00001855139,0.00002958299,0.000110302,0.0004280554,0.000004124202,0.0001859971,0.002210233,0.05208311,0.9442139],"study_design_scores_gemma":[0.001074377,0.0003205574,0.02026471,0.0003027392,0.0000671983,0.00007532038,0.0009543234,0.895548,0.0004617035,0.000237879,0.07958547,0.001107738],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05486906,0.0002083934,0.8605431,0.02536347,0.00180343,0.0008983171,0.003645715,0.001645444,0.05102302],"genre_scores_gemma":[0.5992321,0.00006302429,0.3954629,0.0002864459,0.0002292966,0.000002752729,0.003162615,0.00001434466,0.00154655],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9431061,"threshold_uncertainty_score":0.99991,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null}]}