{"id":"W3197834140","doi":"10.1145/3471158.3472261","title":"The Simplest Thing That Can Possibly Work: (Pseudo-)Relevance Feedback via Text Classification","year":2021,"lang":"en","type":"article","venue":"","topic":"Image Retrieval and Classification Techniques","field":"Computer Science","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Canada First Research Excellence Fund","keywords":"Computer science; Rotation formalisms in three dimensions; Relevance feedback; Relevance (law); Robustness (evolution); Information retrieval; Classifier (UML); Artificial intelligence; Machine learning; Economic shortage; Natural language processing; Data science; Mathematics","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01083226,0.00136539,0.001715249,0.002339949,0.001585676,0.002398128,0.002423715,0.002620649,0.003218669],"category_scores_gemma":[0.04921865,0.000579176,0.001028149,0.001682866,0.002899977,0.007827168,0.002119408,0.002874858,0.00341774],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008415771,"about_ca_system_score_gemma":0.001686345,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001512529,"about_ca_topic_score_gemma":0.002295588,"domain_scores_codex":[0.9878203,0.007562072,0.0004395768,0.001462097,0.002430362,0.0002854838],"domain_scores_gemma":[0.9712559,0.01657995,0.001633297,0.006629101,0.003526143,0.0003754934],"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":[0.001009633,0.0004775868,0.002635995,0.001349643,0.0001990848,0.0001851917,0.0008811508,0.01848087,0.01993986,0.05141769,0.02292454,0.8804988],"study_design_scores_gemma":[0.0004914697,0.002001956,0.003159886,0.0006206698,0.0002656379,0.002142856,0.000535666,0.5173821,0.03674872,0.3683724,0.06776861,0.0005100162],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01424393,0.002932636,0.971759,0.003723365,0.0004498049,0.000268633,0.0001531435,0.00264122,0.003828378],"genre_scores_gemma":[0.3957965,0.001383228,0.5929053,0.002240153,0.001328475,0.0004754267,0.0004132888,0.0003785924,0.005079042],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01083226,"threshold_uncertainty_score":0.05728716,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.034122699067712,"score_gpt":0.2632294582532885,"score_spread":0.2291067591855765,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}