{"id":"W2355952488","doi":"","title":"A Petroleum Subject Collection Strategy Based on the Word co-occurrence Model and the DOM","year":2008,"lang":"en","type":"article","venue":"Microcomputer applications","topic":"Advanced Computational Techniques and Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; Word (group theory); Categorization; Character (mathematics); Tree (set theory); Subject (documents); Artificial intelligence; Natural language processing; Tree structure; Information retrieval; Data structure; Linguistics; World Wide Web","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001331123,0.001120203,0.001111623,0.005737924,0.001750023,0.002721685,0.001340118,0.00106059,0.001602396],"category_scores_gemma":[0.003506438,0.0005121004,0.001486596,0.004360564,0.0009096239,0.006747233,0.001650197,0.001029696,0.001703006],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009734808,"about_ca_system_score_gemma":0.002811971,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008600109,"about_ca_topic_score_gemma":0.009523165,"domain_scores_codex":[0.9981949,0.0004461809,0.0001843412,0.0004452754,0.0006181059,0.000111251],"domain_scores_gemma":[0.9981437,0.0004781106,0.0001452435,0.0004102074,0.0006843931,0.0001383624],"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":[0.0004443434,0.0006065142,0.0128013,0.000714003,0.0001952701,0.0006080571,0.001972103,0.02098648,0.0382516,0.08384581,0.01108896,0.8284856],"study_design_scores_gemma":[0.0001057468,0.0004913462,0.01011096,0.0002031972,0.0004395612,0.002971198,0.001976942,0.715245,0.07446829,0.09688666,0.09667796,0.00042326],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01396227,0.0002853116,0.9806129,0.0001399499,0.00005131538,0.0002817629,0.0002363404,0.00166146,0.002768633],"genre_scores_gemma":[0.137412,0.0004701475,0.8551256,0.0001286102,0.0000760433,0.000365883,0.001277816,0.0003797165,0.004764093],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.008600109,"threshold_uncertainty_score":0.0171001,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02326029468827858,"score_gpt":0.2689521888825048,"score_spread":0.2456918941942262,"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."}}