{"id":"W2048476252","doi":"10.1016/j.ins.2011.03.007","title":"Modeling term proximity for probabilistic information retrieval models","year":2011,"lang":"en","type":"article","venue":"Information Sciences","topic":"Information Retrieval and Search Behavior","field":"Computer Science","cited_by":89,"is_retracted":false,"has_abstract":false,"ca_institutions":"York University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Term (time); Divergence-from-randomness model; Computer science; Probabilistic logic; Term Discrimination; Probabilistic relevance model; Information retrieval; Data mining; Artificial intelligence; Probabilistic analysis of algorithms; Search engine; Concept search","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":[0.005898091,0.001027026,0.002293184,0.003303752,0.001147287,0.00320889,0.002998248,0.00384017,0.004164561],"category_scores_gemma":[0.03849712,0.001445169,0.002332366,0.004070184,0.001351964,0.007560414,0.00218837,0.002864847,0.001634522],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002292605,"about_ca_system_score_gemma":0.001358061,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01044759,"about_ca_topic_score_gemma":0.009870564,"domain_scores_codex":[0.9963197,0.001810089,0.0002326552,0.0008022539,0.0005636772,0.0002716021],"domain_scores_gemma":[0.9819687,0.01517652,0.001105151,0.0008263593,0.0006730899,0.0002501177],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0004184147,0.0002523196,0.004332437,0.0003724401,0.0002645561,0.0002250525,0.0006084892,0.7971791,0.001694063,0.1360554,0.0026961,0.05590159],"study_design_scores_gemma":[0.00002311183,0.00003325273,0.0003261711,0.00001324776,0.00004668949,0.00006147779,0.00002030906,0.9468389,0.0001738649,0.05201257,0.0004318588,0.00001854816],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.08041908,0.002102051,0.9119908,0.001415859,0.0001097001,0.0001543004,0.0005250617,0.0005987241,0.002684362],"genre_scores_gemma":[0.8712137,0.002251678,0.1140756,0.0002974476,0.000397913,0.0005608755,0.001025803,0.0002084648,0.00996851],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01044759,"threshold_uncertainty_score":0.03119242,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1286481222828739,"score_gpt":0.2964142241644975,"score_spread":0.1677661018816236,"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."}}