{"id":"W2051101123","doi":"10.1145/1031171.1031182","title":"Scoring missing terms in information retrieval tasks","year":2004,"lang":"en","type":"article","venue":"","topic":"Information Retrieval and Search Behavior","field":"Computer Science","cited_by":31,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Ministério da Educação; Coordenação de Aperfeiçoamento de Pessoal de Nível Superior","keywords":"Computer science; Query expansion; Information retrieval; Vocabulary; Web query classification; Web search query; Ranking (information retrieval); Query language; Query optimization; Focus (optics); RDF query language; Sargable; Natural language processing; Search engine; Artificial intelligence","routes":{"ca_aff":true,"ca_fund":false,"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":[0.007066137,0.001361258,0.00339286,0.003713154,0.001127616,0.001725019,0.002240524,0.002648487,0.002120689],"category_scores_gemma":[0.04143332,0.0006498481,0.0009768147,0.004275731,0.001179566,0.007746911,0.002098651,0.001686238,0.001430636],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008247521,"about_ca_system_score_gemma":0.0009489359,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004977105,"about_ca_topic_score_gemma":0.00338731,"domain_scores_codex":[0.9927996,0.003193797,0.0007538112,0.000873479,0.002089968,0.0002893619],"domain_scores_gemma":[0.9791632,0.01563678,0.0009375751,0.001835244,0.001943303,0.0004838463],"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":[0.002982598,0.000776983,0.0131762,0.001911501,0.0003455607,0.0005468655,0.001368652,0.08308536,0.04519261,0.009016669,0.008275238,0.8333218],"study_design_scores_gemma":[0.0003764712,0.001463301,0.009704445,0.0001230068,0.000463756,0.0009829173,0.0006386499,0.8835329,0.0481503,0.04593823,0.008400625,0.0002253574],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2528192,0.01334404,0.7260681,0.000678231,0.0002413864,0.0004995018,0.0005767462,0.003546845,0.002225987],"genre_scores_gemma":[0.596754,0.001976534,0.39504,0.0002047807,0.0003543682,0.0003267764,0.002027037,0.0002898687,0.00302653],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.007066137,"threshold_uncertainty_score":0.03736973,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01795538351522202,"score_gpt":0.2604083802063343,"score_spread":0.2424529966911123,"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."}}