{"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","author_layer_release":"2026-06-26"},"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","authors":[{"name":"Trevor N. Mansuy","is_ca":true},{"name":"Robert J. Hilderman","is_ca":true}],"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00345881,0.001159865,0.001044076,0.001683487,0.0007139039,0.001869194,0.001196253,0.001274838,0.002615798],"category_scores_gemma":[0.01821428,0.0004477136,0.0009100479,0.001922632,0.001106071,0.006006077,0.000780198,0.001565477,0.000879532],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001198635,"about_ca_system_score_gemma":0.0007434677,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002489662,"about_ca_topic_score_gemma":0.004628894,"domain_scores_codex":[0.998337,0.0006174144,0.0001273256,0.0003192053,0.0004800511,0.0001188795],"domain_scores_gemma":[0.9888839,0.008735802,0.000800522,0.0007328872,0.0006994359,0.0001474464],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0008311489,0.000488301,0.01402611,0.000584554,0.0001493087,0.0004527931,0.0006982604,0.4118333,0.02222948,0.1439622,0.004971245,0.3997733],"study_design_scores_gemma":[0.00001206729,0.00008919649,0.000695685,0.00002309078,0.00001710281,0.0001257004,0.00002514001,0.9544361,0.001620481,0.0419467,0.0009937567,0.00001503062],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05810428,0.0003987664,0.9370002,0.0006399466,0.00006673665,0.0001630052,0.0004368623,0.0008689995,0.002321188],"genre_scores_gemma":[0.7763625,0.0004791951,0.2175921,0.0003433058,0.0001782597,0.0005685643,0.0009798359,0.000179695,0.003316609],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.00345881,"threshold_uncertainty_score":0.01829219,"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","authors":[{"name":"Jian Pei","is_ca":true}],"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01003651,0.001220451,0.002984072,0.004561246,0.0009197942,0.004947164,0.002764899,0.002516437,0.001040821],"category_scores_gemma":[0.04024334,0.001583882,0.002387036,0.006794104,0.001662759,0.009602892,0.003852092,0.006941668,0.00186048],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001297401,"about_ca_system_score_gemma":0.002027587,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002675886,"about_ca_topic_score_gemma":0.002986814,"domain_scores_codex":[0.9940436,0.002425056,0.0004136929,0.001067979,0.001860364,0.0001893493],"domain_scores_gemma":[0.9732766,0.01633543,0.001176,0.005349915,0.003195586,0.0006663943],"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":[0.0004594266,0.0014805,0.0293846,0.001874151,0.001035428,0.0009538346,0.001577031,0.2198498,0.01657764,0.2293893,0.042003,0.4554153],"study_design_scores_gemma":[0.00005921153,0.00005522839,0.00126737,0.00008581197,0.00004755408,0.0001505737,0.0001984122,0.7455148,0.002116469,0.2371021,0.01336602,0.00003651146],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02247662,0.004302252,0.9592943,0.008032251,0.0001887101,0.0002464259,0.0008455352,0.002752846,0.001861087],"genre_scores_gemma":[0.1144885,0.003904024,0.875137,0.001634346,0.0007722658,0.0005127362,0.001775442,0.0002788544,0.001496857],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01003651,"threshold_uncertainty_score":0.05307877,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null}]}